Topic hub
Does the bitterness of PROP and PTC differ by TAS2R38 genotype? (Are supertasters the same as a genotype?)
Sensory tests in which groups of people with different genotypes rated standard PROP and PTC solutions, impregnated paper and dissolvable films
To be clear from the start, this page is about studies in which groups of people with different types of the bitter-receptor gene TAS2R38 rated PROP (propylthiouracil) and PTC (phenylthiocarbamide), compounds used as standard stimuli in bitterness research, as solutions, impregnated paper or dissolvable films. The types are mainly three combinations (diplotypes): PAV/PAV, PAV/AVI and AVI/AVI. The values compared in the tables are group bitterness ratings (means, medians or geometric means), thresholds expressed as solution concentrations, and the dilution number at which bitterness was recognised, as stated row by row in the unit column. Studies read in full that give no group values, only regression coefficients or an order stated in the text, and a study giving only the share of people answering bitter at the same concentration, are not tabulated and are described in the reading sentences and the scope table.
This page does not sort people by genotype to decide their food preferences or health, and it does not recommend genetic testing. Even where group averages differ, ratings are widely spread within the same genotype, and how each person perceives bitterness is not decided by the group values.
Studies comparing the bitterness of brassica vegetables by genotype (those included in full are on rocket and turnip), and sources on vegetable compounds, are on the page Is finding broccoli, kale and other brassicas bitter down to a gene (TAS2R38)?. The line is drawn by whether the main sample compared is a standard PROP or PTC stimulus or a vegetable or food. This page does not tabulate studies that used vegetables or foods as samples.
"Supertaster" is a name for people who rated the bitterness of a PROP solution as very strong; it is not the name of a genotype. When groups are defined by the same ratings that are then compared, the difference largely follows from the definition. When ratings measured on another day or with another presentation than the classification are compared, the result is not fixed by the definition alone, but because the classification uses PROP ratings it is not treated as evidence independent of genotype. For this reason the pairs in the tables are only between groups defined by genotype, and an example in which rating-based groups and genotype groups overlapped but did not match is given in the reading sentences.
Presentation varies: solutions, impregnated paper (filter-paper disks or strips soaked in solution) and dissolvable films. For impregnated paper, the concentration is that of the soaking solution, not the concentration in the mouth; for films, the amount is the content of one film. Measures also vary: bitterness intensity ratings, thresholds expressed as solution concentrations, and dilution numbers. Each is stated in the condition and unit columns.
Many studies give values only in figures, so please check, row by row, where each value comes from, the test line and the notes. Scales, summaries, presentation and populations differ between studies, so values are not compared or turned into ratios across studies. With that in mind, the page separates what was established within one experiment, candidate explanations from differences between studies, and what this comparison alone cannot separate.
On this page (9 sections)
- What this page establishes
- What the included studies allow us to say
- Bitterness ratings at a glance
- Conditions and results, study by study
- What was established within an experiment
- Candidate explanations for the differences
- What is not known — what this comparison alone cannot separate
- How studies and rows were counted
- What we looked at
What this page establishes
- It lines up, study by study and measure by measure, studies in which genotype groups rated standard PROP or PTC stimuli, with presentation, concentration, and who rated on what scale.
- The tabulated studies agreed in direction: in pairs with numeric group intensity ratings the AVI/AVI groups had lower ratings, and for thresholds expressed as concentrations the AVI/AVI groups had higher concentrations.
- Whether PAV/PAV and PAV/AVI differed split by study, presentation and measure into cases with and without a significant difference; no significant difference does not mean the same.
- "Supertaster" groups defined by PROP ratings overlapped with genotype groups but did not match them.
- Differences in group averages do not decide how each person perceives bitterness, or their food preferences or health; nothing can be said from this table about the bitterness of vegetables or foods.
What the included studies allow us to say
- Among studies in which groups of people with different TAS2R38 genotypes rated standard PROP or PTC stimuli (solutions, paper soaked in solution, or dissolvable films), eleven were checked in full text and tabulated, giving fifty compared pairs. Three more read in full give no group values or cannot be compared as concentration thresholds, so they are not tabulated and are described in the answer below and in the scope table. Three of the eleven tabulated studies come from the same laboratory (University of Cagliari, Italy) and are not counted as independent studies. Samples, presentation, scales and populations differ between studies, so values are not compared across studies.
- The direction was consistent. In every pair with numeric group intensity ratings, the published value for the AVI/AVI group was lower than for the PAV/PAV and PAV/AVI groups. For thresholds expressed as concentrations, the AVI/AVI group had higher concentrations, needing a stronger solution to detect it. The dilution number at which bitterness was recognised was also lower for the AVI/AVI group. However, pairs with pairwise tests, pairs with only an overall test, and pairs whose values are approximate readings from figures are mixed. [rows: Fischer 2014 Paper disk, PAV/PAV vs AVI/AVI; Fischer 2014 Paper disk, PAV/AVI vs AVI/AVI; Duffy 2004 3.2 mM, PAV/PAV vs AVI/AVI; Duffy 2004 3.2 mM, PAV/AVI vs AVI/AVI; Melis 2015 3.2 mM, PAV/PAV vs AVI/AVI; Melis 2015 3.2 mM, PAV/AVI vs AVI/AVI; Aoki 2023 Threshold, PAV/PAV vs AVI/AVI; Aoki 2023 Threshold, PAV/AVI vs AVI/AVI; Bufe 2005 PTC recognition threshold, PAV/PAV vs AVI/AVI; Bufe 2005 PTC recognition threshold, PAV/AVI vs AVI/AVI; Campbell 2012 PTC dilution score, PAV/PAV vs AVI/AVI]
- Whether PAV/PAV and PAV/AVI differed depended on the study and the measure. Some pairs showed significant differences (the filter-paper disk, the 3.2 mM solution of Duffy and colleagues, the thresholds of Aoki and colleagues, and others), and some did not (the solution of Melis and colleagues in 2015, the paper disk of Sollai and colleagues, the PTC recognition threshold of Bufe and colleagues, the film of Smutzer and colleagues, and others). No significant difference does not mean the same. [rows: Fischer 2014 Paper disk, PAV/PAV vs PAV/AVI; Duffy 2004 3.2 mM, PAV/PAV vs PAV/AVI; Aoki 2023 Threshold, PAV/PAV vs PAV/AVI; Melis 2015 3.2 mM, PAV/PAV vs PAV/AVI; Sollai 2017 Paper disk, PAV/PAV vs PAV/AVI; Bufe 2005 PTC recognition threshold, PAV/PAV vs PAV/AVI; Smutzer 2013 Film 600 nmol, PAV/PAV vs PAV/AVI]
- Differences in group averages did not decide how each person perceived bitterness. Ratings were widely spread within the same genotype, and some PAV/PAV people rated very low while some AVI/AVI people rated very high. This page does not sort people by genotype to decide their food preferences or health, and it does not recommend genetic testing. [rows: Fischer 2014 Paper disk, PAV/PAV vs AVI/AVI; Bufe 2005 PROP 5500 µM, PAV/PAV vs AVI/AVI; Nolden 2024 3.2 mM, PAV/PAV vs AVI/AVI]
- The three studies not tabulated point the same way. Garneau and colleagues give no group rating values; they tested adjusted regression coefficients with AVI/AVI as the reference, and the coefficients for PAV/PAV and PAV/AVI were both positive and significant. These are not the group mean ratings themselves. A t-test of the PAV-carrying groups combined against AVI/AVI was also significant. Behrens and colleagues give no table of group thresholds; the text says PAV/PAV thresholds were lowest and AVI/AVI highest, and a regression using the number of PAV alleles carried showed a significant association with threshold. In Mennella and colleagues, the share answering bitter at the weakest solution was highest for PAV/PAV and lowest for AVI/AVI, but this is a share answering at the same concentration and does not show which group has the lower concentration threshold. Values and p values are in the scope table.
Bitterness ratings at a glance
Only results of people's bitterness ratings are extracted here. Presentation, scales and summaries differ between studies, so values are not compared across studies. Each row is a pair of two genotype groups: the genotype written before "vs" in the heading is the "genotype group named first" and the one after it the "genotype group named second", and the value columns follow the same order. The direction column shows, for bitterness intensity rows, the group with the larger value as "… more bitter". Rows measured as "bitterness threshold" show "… detected bitterness at lower concentrations": the group with the smaller value for concentration thresholds, and the larger value for the dilution number at which bitterness was recognised (a larger number is a more dilute solution). Shares of people answering bitter at the same concentration do not show which group has the lower concentration threshold, so they are not tabulated. Direction is set from the size of the published values. Whether a difference is significant is shown in the test line of the "Direction / test" column. Rows with only an overall test that does not say which pair differs are not described as showing a significant difference for that pair. Rows where the test method is described but no result is given for the pair shown are marked "result for this comparison not reported". Rows whose values exist only in a figure say in the unit column that they are approximate readings. Click a study name to jump to its detailed table under "Conditions and results, study by study".
Results in which people rated bitterness only. Where each value comes from (a table or a reading from a figure) is shown in its row; scales and methods differ between studies, so compare within a row, not across rows.
| Study | Comparison and condition | Measured as | Genotype group named first | Genotype group named second | Direction / test |
|---|---|---|---|---|---|
| Bufe 2005 | PTC recognition threshold, PAV/PAV vs AVI/AVI PTC in water; recognition threshold by a modified Harris–Kalmus method | recognition threshold concentration (µM, group mean; lower means bitterness recognised in a more dilute solution) | 25.0 µM | 4730 µM | Genotype group named first detected bitterness at lower concentrations Tested, significant |
| Bufe 2005 | PTC recognition threshold, PAV/AVI vs AVI/AVI PTC in water; recognition threshold by a modified Harris–Kalmus method | recognition threshold concentration (µM, group mean; lower means bitterness recognised in a more dilute solution) | 40.6 µM | 4730 µM | Genotype group named first detected bitterness at lower concentrations Tested, significant |
| Bufe 2005 | PTC recognition threshold, PAV/PAV vs PAV/AVI PTC in water; recognition threshold by a modified Harris–Kalmus method | recognition threshold concentration (µM, group mean; lower means bitterness recognised in a more dilute solution) | 25.0 µM | 40.6 µM | Genotype group named first detected bitterness at lower concentrations Tested, not significant |
| Bufe 2005 | PROP recognition threshold, PAV/PAV vs AVI/AVI PROP in water; recognition threshold by a modified Harris–Kalmus method | recognition threshold concentration (µM, group mean; lower means bitterness recognised in a more dilute solution) | 58.0 µM | 1410 µM | Genotype group named first detected bitterness at lower concentrations Tested; significance reported, but which pair differs is unclear |
| Bufe 2005 | PROP recognition threshold, PAV/AVI vs AVI/AVI PROP in water; recognition threshold by a modified Harris–Kalmus method | recognition threshold concentration (µM, group mean; lower means bitterness recognised in a more dilute solution) | 217.0 µM | 1410 µM | Genotype group named first detected bitterness at lower concentrations Tested; significance reported, but which pair differs is unclear |
| Bufe 2005 | PROP recognition threshold, PAV/PAV vs PAV/AVI PROP in water; recognition threshold by a modified Harris–Kalmus method | recognition threshold concentration (µM, group mean; lower means bitterness recognised in a more dilute solution) | 58.0 µM | 217.0 µM | Genotype group named first detected bitterness at lower concentrations Tested; significance reported, but which pair differs is unclear |
| Bufe 2005 | PTC 5600 µM, PAV/PAV vs AVI/AVI PTC 5600 µM in water (the strongest solution in Fig. 3) | bitterness intensity (group geometric mean of standardized gLMS ratings; higher is stronger); approximate values from the small Fig. 3 insets, rounded to the precision that can be read | about 60 | about 10–15 | Genotype group named first more bitter Tested, significant |
| Bufe 2005 | PTC 5600 µM, PAV/AVI vs AVI/AVI PTC 5600 µM in water (the strongest solution in Fig. 3) | bitterness intensity (group geometric mean of standardized gLMS ratings; higher is stronger); approximate values from the small Fig. 3 insets, rounded to the precision that can be read | about 40 | about 10–15 | Genotype group named first more bitter Tested, significant |
| Bufe 2005 | PTC 5600 µM, PAV/PAV vs PAV/AVI PTC 5600 µM in water (the strongest solution in Fig. 3) | bitterness intensity (group geometric mean of standardized gLMS ratings; higher is stronger); approximate values from the small Fig. 3 insets, rounded to the precision that can be read | about 60 | about 40 | Genotype group named first more bitter Tested, significant |
| Bufe 2005 | PROP 550 µM, PAV/PAV vs AVI/AVI PROP 550 µM in water (the third concentration in Fig. 3) | bitterness intensity (group geometric mean of standardized gLMS ratings; higher is stronger); approximate values from the small Fig. 3 insets, rounded to the precision that can be read | about 12 | about 1–2 | Genotype group named first more bitter Tested, significant |
| Bufe 2005 | PROP 550 µM, PAV/AVI vs AVI/AVI PROP 550 µM in water (the third concentration in Fig. 3) | bitterness intensity (group geometric mean of standardized gLMS ratings; higher is stronger); approximate values from the small Fig. 3 insets, rounded to the precision that can be read | about 3 | about 1–2 | Genotype group named first more bitter Test method described; result for this comparison not reported |
| Bufe 2005 | PROP 550 µM, PAV/PAV vs PAV/AVI PROP 550 µM in water (the third concentration in Fig. 3) | bitterness intensity (group geometric mean of standardized gLMS ratings; higher is stronger); approximate values from the small Fig. 3 insets, rounded to the precision that can be read | about 12 | about 3 | Genotype group named first more bitter Tested, significant |
| Bufe 2005 | PROP 5500 µM, PAV/PAV vs AVI/AVI PROP 5500 µM in water (the strongest solution in Fig. 3) | bitterness intensity (group geometric mean of standardized gLMS ratings; higher is stronger); approximate values from the small Fig. 3 insets, rounded to the precision that can be read | about 21 | about 12 | Genotype group named first more bitter Test method described; result for this comparison not reported |
| Duffy 2004 | Threshold, PAV/PAV vs AVI/AVI PROP in water; up-down threshold (the first 48 participants) | threshold concentration (mM, mean ± SEM; lower means detected in a more dilute solution) | 0.011 ± 0.003 mM | 0.579 ± 0.10 mM | Genotype group named first detected bitterness at lower concentrations Tested; significance reported, but which pair differs is unclear |
| Duffy 2004 | Threshold, PAV/AVI vs AVI/AVI PROP in water; up-down threshold (the first 48 participants) | threshold concentration (mM, mean ± SEM; lower means detected in a more dilute solution) | 0.038 ± 0.001 mM | 0.579 ± 0.10 mM | Genotype group named first detected bitterness at lower concentrations Tested; significance reported, but which pair differs is unclear |
| Duffy 2004 | Threshold, PAV/PAV vs PAV/AVI PROP in water; up-down threshold (the first 48 participants) | threshold concentration (mM, mean ± SEM; lower means detected in a more dilute solution) | 0.011 ± 0.003 mM | 0.038 ± 0.001 mM | Genotype group named first detected bitterness at lower concentrations Tested; significance reported, but which pair differs is unclear |
| Duffy 2004 | 3.2 mM, PAV/PAV vs AVI/AVI PROP 3.2 mM in water (the strongest of 5 concentrations) | bitterness intensity (gLMS 0–100, group mean; higher is stronger); approximate values from Fig. 2 | about 50 | about 20 | Genotype group named first more bitter Tested, significant |
| Duffy 2004 | 3.2 mM, PAV/AVI vs AVI/AVI PROP 3.2 mM in water (the strongest of 5 concentrations) | bitterness intensity (gLMS 0–100, group mean; higher is stronger); approximate values from Fig. 2 | about 40 | about 20 | Genotype group named first more bitter Tested, significant |
| Duffy 2004 | 3.2 mM, PAV/PAV vs PAV/AVI PROP 3.2 mM in water (the strongest of 5 concentrations) | bitterness intensity (gLMS 0–100, group mean; higher is stronger); approximate values from Fig. 2 | about 50 | about 40 | Genotype group named first more bitter Tested, significant |
| Fischer 2014 | Paper disk, PAV/PAV vs AVI/AVI Filter-paper disk impregnated with 1.0 M PROP (concentration of the soaking solution) | bitterness intensity (gLMS 0–100, group mean; higher is stronger) | 60.1 | 14.4 | Genotype group named first more bitter Tested, significant |
| Fischer 2014 | Paper disk, PAV/AVI vs AVI/AVI Filter-paper disk impregnated with 1.0 M PROP (concentration of the soaking solution) | bitterness intensity (gLMS 0–100, group mean; higher is stronger) | 46.5 | 14.4 | Genotype group named first more bitter Tested, significant |
| Fischer 2014 | Paper disk, PAV/PAV vs PAV/AVI Filter-paper disk impregnated with 1.0 M PROP (concentration of the soaking solution) | bitterness intensity (gLMS 0–100, group mean; higher is stronger) | 60.1 | 46.5 | Genotype group named first more bitter Tested, significant |
| Melis 2013 | Threshold, PAV/PAV vs AVI/AVI PROP in water; ascending 3-AFC threshold | threshold concentration (mM, group mean; lower means detected in a more dilute solution); approximate values from Fig. 1 | about 0.2 mM | about 2.8 mM | Genotype group named first detected bitterness at lower concentrations Tested, significant |
| Melis 2013 | Threshold, PAV/AVI vs AVI/AVI PROP in water; ascending 3-AFC threshold | threshold concentration (mM, group mean; lower means detected in a more dilute solution); approximate values from Fig. 1 | about 0.9 mM | about 2.8 mM | Genotype group named first detected bitterness at lower concentrations Tested, significant |
| Melis 2013 | Threshold, PAV/PAV vs PAV/AVI PROP in water; ascending 3-AFC threshold | threshold concentration (mM, group mean; lower means detected in a more dilute solution); approximate values from Fig. 1 | about 0.2 mM | about 0.9 mM | Genotype group named first detected bitterness at lower concentrations Tested, not significant |
| Melis 2013 | 3.2 mM, PAV/PAV vs AVI/AVI 10 mL of 3.2 mM PROP in water for 10 s | bitterness intensity (LMS, group mean; higher is stronger); approximate values from Fig. 1 | about 77 | about 24.5 | Genotype group named first more bitter Tested, significant |
| Melis 2013 | 3.2 mM, PAV/AVI vs AVI/AVI 10 mL of 3.2 mM PROP in water for 10 s | bitterness intensity (LMS, group mean; higher is stronger); approximate values from Fig. 1 | about 61.5 | about 24.5 | Genotype group named first more bitter Tested, significant |
| Melis 2013 | 3.2 mM, PAV/PAV vs PAV/AVI 10 mL of 3.2 mM PROP in water for 10 s | bitterness intensity (LMS, group mean; higher is stronger); approximate values from Fig. 1 | about 77 | about 61.5 | Genotype group named first more bitter Tested, significant |
| Melis 2015 | 3.2 mM, PAV/PAV vs AVI/AVI 4 mL of 3.2 mM PROP in water (nothing added) | bitterness intensity (LMS 0–100, group mean; higher is stronger); approximate values from Fig. 4A | about 69 | about 3 | Genotype group named first more bitter Tested, significant |
| Melis 2015 | 3.2 mM, PAV/AVI vs AVI/AVI 4 mL of 3.2 mM PROP in water (nothing added) | bitterness intensity (LMS 0–100, group mean; higher is stronger); approximate values from Fig. 4A | about 51 | about 3 | Genotype group named first more bitter Tested, significant |
| Melis 2015 | 3.2 mM, PAV/PAV vs PAV/AVI 4 mL of 3.2 mM PROP in water (nothing added) | bitterness intensity (LMS 0–100, group mean; higher is stronger); approximate values from Fig. 4A | about 69 | about 51 | Genotype group named first more bitter Tested, not significant |
| Sollai 2017 | Paper disk, PAV/PAV vs AVI/AVI Paper disk with 30 µL of 3.2 mM PROP on the tongue tip for 15 s | bitterness intensity (LMS, group median; higher is stronger); approximate values from Fig. 1 | about 60 | about 3 | Genotype group named first more bitter Tested, significant |
| Sollai 2017 | Paper disk, PAV/AVI vs AVI/AVI Paper disk with 30 µL of 3.2 mM PROP on the tongue tip for 15 s | bitterness intensity (LMS, group median; higher is stronger); approximate values from Fig. 1 | about 35 | about 3 | Genotype group named first more bitter Tested, significant |
| Sollai 2017 | Paper disk, PAV/PAV vs PAV/AVI Paper disk with 30 µL of 3.2 mM PROP on the tongue tip for 15 s | bitterness intensity (LMS, group median; higher is stronger); approximate values from Fig. 1 | about 60 | about 35 | Genotype group named first more bitter Tested, not significant |
| Nolden 2024 | 3.2 mM, PAV/PAV vs AVI/AVI PROP 3.2 mM in water (the strongest of 5 concentrations) | bitterness intensity (gLMS, group mean; higher is stronger); approximate values from Fig. 3, converted using the scale positions stated in the 2020 paper from the same study | about 55 | about 20 | Genotype group named first more bitter Tested, significant |
| Nolden 2024 | 3.2 mM, PAV/AVI vs AVI/AVI PROP 3.2 mM in water (the strongest of 5 concentrations) | bitterness intensity (gLMS, group mean; higher is stronger); approximate values from Fig. 3, converted using the scale positions stated in the 2020 paper from the same study | about 44 | about 20 | Genotype group named first more bitter Tested, significant |
| Nolden 2024 | 3.2 mM, PAV/PAV vs PAV/AVI PROP 3.2 mM in water (the strongest of 5 concentrations) | bitterness intensity (gLMS, group mean; higher is stronger); approximate values from Fig. 3, converted using the scale positions stated in the 2020 paper from the same study | about 55 | about 44 | Genotype group named first more bitter Tested, significant |
| Aoki 2023 | Threshold, PAV/PAV vs AVI/AVI 10 mL of PROP in water; detection threshold by modified 2AFC with QUEST | detection threshold concentration (mM, group median; lower means detected in a more dilute solution); approximate values from Fig. 4 (log axis) | about 0.025 mM | about 1.3 mM | Genotype group named first detected bitterness at lower concentrations Tested, significant |
| Aoki 2023 | Threshold, PAV/AVI vs AVI/AVI 10 mL of PROP in water; detection threshold by modified 2AFC with QUEST | detection threshold concentration (mM, group median; lower means detected in a more dilute solution); approximate values from Fig. 4 (log axis) | about 0.043 mM | about 1.3 mM | Genotype group named first detected bitterness at lower concentrations Tested, significant |
| Aoki 2023 | Threshold, PAV/PAV vs PAV/AVI 10 mL of PROP in water; detection threshold by modified 2AFC with QUEST | detection threshold concentration (mM, group median; lower means detected in a more dilute solution); approximate values from Fig. 4 (log axis) | about 0.025 mM | about 0.043 mM | Genotype group named first detected bitterness at lower concentrations Tested, significant |
| Campbell 2012 | PTC dilution score, PAV/PAV vs AVI/AVI PTC dilution series (13 most dilute, 1 most concentrated); recognition method in the field | number of the most dilute solution at which bitterness was recognised (group mean; higher means recognised in a more dilute solution) | 9.51 ± 1.66 | 3.36 ± 2.11 | Genotype group named first detected bitterness at lower concentrations Tested; significance reported, but which pair differs is unclear |
| Graham 2021 | Strip, PAV/PAV vs AVI/AVI Commercial PTC-impregnated strip (amount and concentration not stated) | bitterness intensity (gLMS 0–100, group median [interquartile range]; higher is stronger) | 31 [30] | 2 [6] | Genotype group named first more bitter Tested, significant |
| Graham 2021 | Strip, PAV/AVI vs AVI/AVI Commercial PTC-impregnated strip (amount and concentration not stated) | bitterness intensity (gLMS 0–100, group median [interquartile range]; higher is stronger) | 9 [24] | 2 [6] | Genotype group named first more bitter Tested, significant |
| Graham 2021 | Strip, PAV/PAV vs PAV/AVI Commercial PTC-impregnated strip (amount and concentration not stated) | bitterness intensity (gLMS 0–100, group median [interquartile range]; higher is stronger) | 31 [30] | 9 [24] | Genotype group named first more bitter Test method described; result for this comparison not reported |
| Smutzer 2013 | Solution 3.2 mM, PAV/PAV vs AVI/AVI 10 mL of 3.2 mM PROP in water | bitterness intensity (gLMS, group mean; higher is stronger); approximate values from Fig. 1B | about 61.5 | about 15 | Genotype group named first more bitter Tested, significant |
| Smutzer 2013 | Solution 3.2 mM, PAV/AVI vs AVI/AVI 10 mL of 3.2 mM PROP in water | bitterness intensity (gLMS, group mean; higher is stronger); approximate values from Fig. 1B | about 50.5 | about 15 | Genotype group named first more bitter Tested, significant |
| Smutzer 2013 | Solution 3.2 mM, PAV/PAV vs PAV/AVI 10 mL of 3.2 mM PROP in water | bitterness intensity (gLMS, group mean; higher is stronger); approximate values from Fig. 1B | about 61.5 | about 50.5 | Genotype group named first more bitter Tested, significant |
| Smutzer 2013 | Film 600 nmol, PAV/PAV vs AVI/AVI Dissolvable film containing 600 nmol PROP (amount in 1 film, not a solution concentration) | bitterness intensity (gLMS, group mean; higher is stronger); approximate values from Fig. 1A | about 43.5 | about 13 | Genotype group named first more bitter Tested, significant |
| Smutzer 2013 | Film 600 nmol, PAV/AVI vs AVI/AVI Dissolvable film containing 600 nmol PROP (amount in 1 film, not a solution concentration) | bitterness intensity (gLMS, group mean; higher is stronger); approximate values from Fig. 1A | about 37 | about 13 | Genotype group named first more bitter Tested, significant |
| Smutzer 2013 | Film 600 nmol, PAV/PAV vs PAV/AVI Dissolvable film containing 600 nmol PROP (amount in 1 film, not a solution concentration) | bitterness intensity (gLMS, group mean; higher is stronger); approximate values from Fig. 1A | about 43.5 | about 37 | Genotype group named first more bitter Tested, not significant |
Conditions and results, study by study
For each study, the sample and presentation, how it was served, and who rated it are listed with the results. A row is "study × measured result × genotype pair"; different measures and different genotype pairs within one study get separate rows. Under each row are its location in the paper and whether it can be used for the comparison "whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus, bitterness intensity or thresholds differ between the groups", with the reason. No cell is left blank: "not stated" (not in the paper), "not read" and "not applicable" are distinguished. For studies whose main subject lies elsewhere (alcohol, fat taste, electrophysiology and so on), only the group comparisons of PROP or PTC ratings are used.
Bufe 2005 sensory only
The molecular basis of individual differences in phenylthiocarbamide and propylthiouracil bitterness perception. Current Biology — doi:10.1016/j.cub.2005.01.047
- How deeply checked
- Checked in full text — Author manuscript on PMC (text, Figs. 3 and 4 and the table under Fig. 4); the insets of Fig. 3 were read by eye from an enlarged image. The supplement with detailed methods was not obtained (numbers per group, volume and holding time not checked). Cell assays of the receptor are the main part of the paper; human ratings are one part Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-24.
- Samples
- PTC and PROP in water. Suprathreshold intensity at the 5 concentrations of Fig. 3 (PTC 56, 180, 560, 1800, 5600 µM; PROP 55, 174, 550, 1740, 5500 µM; the text gives the second and third PROP concentrations as 170 and 550 µM, which differs from the figure). Recognition thresholds by a modified Harris–Kalmus method
- Processing and cooking
- Not applicable (solutions only)
- How it was served or eaten
- Not stated (volume and holding time are not in the text; the supplement was not obtained)
- Who rated, on what scale
- 32 genotyped people (PAV/PAV, PAV/AVI, AVI/AVI; numbers per group not in the text). Intensity on the gLMS; Fig. 3 shows standardized ratings and the insets show group geometric means. A further 3 people (1 each of PAV/PAV, AVI/AVI, AAI/AAI) were tested in detail but are not given rows because each is one person
| Measured | Comparison and condition | Genotype group named first | Genotype group named second | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness threshold (sensory) | PTC recognition threshold, PAV/PAV vs AVI/AVI PTC in water; recognition threshold by a modified Harris–Kalmus method | 25.0 µM | 4730 µM | recognition threshold concentration (µM, group mean; lower means bitterness recognised in a more dilute solution) | Genotype group named first detected bitterness at lower concentrations Tested, significant |
| Group difference F[2,27]=20.72, p<0.0001; AVI/AVI significantly higher than the other 2 groups (p<0.001) Where: Results, 'Recognition Threshold Measures'; Fig. 4 and the table beneath it For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Usable — Genotype groups rated the same standard solution; numbers per group are not in the text, and volume and holding time are in a supplement that was not checked | |||||
| Bitterness threshold (sensory) | PTC recognition threshold, PAV/AVI vs AVI/AVI PTC in water; recognition threshold by a modified Harris–Kalmus method | 40.6 µM | 4730 µM | recognition threshold concentration (µM, group mean; lower means bitterness recognised in a more dilute solution) | Genotype group named first detected bitterness at lower concentrations Tested, significant |
| Group difference F[2,27]=20.72, p<0.0001; AVI/AVI significantly higher than the other 2 groups (p<0.001) Where: Results, 'Recognition Threshold Measures'; Fig. 4 and the table beneath it For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Usable — Genotype groups rated the same standard solution; numbers per group are not in the text, and volume and holding time are in a supplement that was not checked | |||||
| Bitterness threshold (sensory) | PTC recognition threshold, PAV/PAV vs PAV/AVI PTC in water; recognition threshold by a modified Harris–Kalmus method | 25.0 µM | 40.6 µM | recognition threshold concentration (µM, group mean; lower means bitterness recognised in a more dilute solution) | Genotype group named first detected bitterness at lower concentrations Tested, not significant |
| The text says PAV/PAV and PAV/AVI did not differ from each other (no p value given); no significant difference does not mean the same Where: Results, 'Recognition Threshold Measures'; Fig. 4 and the table beneath it For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Usable — Genotype groups rated the same standard solution; numbers per group are not in the text, and volume and holding time are in a supplement that was not checked | |||||
| Bitterness threshold (sensory) | PROP recognition threshold, PAV/PAV vs AVI/AVI PROP in water; recognition threshold by a modified Harris–Kalmus method | 58.0 µM | 1410 µM | recognition threshold concentration (µM, group mean; lower means bitterness recognised in a more dilute solution) | Genotype group named first detected bitterness at lower concentrations Tested; significance reported, but which pair differs is unclear |
| Group difference F[2,25]=7.95, p<0.01; the text only says "similar differences" occurred with PROP as with PTC (p<0.01) and does not name which pairs differ for PROP Where: Results, 'Recognition Threshold Measures'; Fig. 4 and the table beneath it For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same standard solution; numbers per group are not in the text, and volume and holding time are in a supplement that was not checked; for PROP, which pairs differ is not stated | |||||
| Bitterness threshold (sensory) | PROP recognition threshold, PAV/AVI vs AVI/AVI PROP in water; recognition threshold by a modified Harris–Kalmus method | 217.0 µM | 1410 µM | recognition threshold concentration (µM, group mean; lower means bitterness recognised in a more dilute solution) | Genotype group named first detected bitterness at lower concentrations Tested; significance reported, but which pair differs is unclear |
| Group difference F[2,25]=7.95, p<0.01; the text only says "similar differences" occurred with PROP as with PTC (p<0.01) and does not name which pairs differ for PROP Where: Results, 'Recognition Threshold Measures'; Fig. 4 and the table beneath it For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same standard solution; numbers per group are not in the text, and volume and holding time are in a supplement that was not checked; for PROP, which pairs differ is not stated | |||||
| Bitterness threshold (sensory) | PROP recognition threshold, PAV/PAV vs PAV/AVI PROP in water; recognition threshold by a modified Harris–Kalmus method | 58.0 µM | 217.0 µM | recognition threshold concentration (µM, group mean; lower means bitterness recognised in a more dilute solution) | Genotype group named first detected bitterness at lower concentrations Tested; significance reported, but which pair differs is unclear |
| Group difference F[2,25]=7.95, p<0.01; the text only says "similar differences" occurred with PROP as with PTC (p<0.01) and does not name which pairs differ for PROP Where: Results, 'Recognition Threshold Measures'; Fig. 4 and the table beneath it For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same standard solution; numbers per group are not in the text, and volume and holding time are in a supplement that was not checked; for PROP, which pairs differ is not stated | |||||
| Bitterness intensity (sensory) | PTC 5600 µM, PAV/PAV vs AVI/AVI PTC 5600 µM in water (the strongest solution in Fig. 3) | about 60 | about 10–15 | bitterness intensity (group geometric mean of standardized gLMS ratings; higher is stronger); approximate values from the small Fig. 3 insets, rounded to the precision that can be read | Genotype group named first more bitter Tested, significant |
| The text says the geometric means in the insets of all 3 groups differed significantly from each other at every PTC concentration (p<0.001). Readings from a figure carry some margin. Individual lines spread widely within groups Where: Results, 'Suprathreshold Measures'; Fig. 3 insets For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same standard solution; numbers per group are not in the text, and volume and holding time are in a supplement that was not checked; values approximate from a figure | |||||
| Bitterness intensity (sensory) | PTC 5600 µM, PAV/AVI vs AVI/AVI PTC 5600 µM in water (the strongest solution in Fig. 3) | about 40 | about 10–15 | bitterness intensity (group geometric mean of standardized gLMS ratings; higher is stronger); approximate values from the small Fig. 3 insets, rounded to the precision that can be read | Genotype group named first more bitter Tested, significant |
| The text says the geometric means in the insets of all 3 groups differed significantly from each other at every PTC concentration (p<0.001). Readings from a figure carry some margin. Individual lines spread widely within groups Where: Results, 'Suprathreshold Measures'; Fig. 3 insets For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same standard solution; numbers per group are not in the text, and volume and holding time are in a supplement that was not checked; values approximate from a figure | |||||
| Bitterness intensity (sensory) | PTC 5600 µM, PAV/PAV vs PAV/AVI PTC 5600 µM in water (the strongest solution in Fig. 3) | about 60 | about 40 | bitterness intensity (group geometric mean of standardized gLMS ratings; higher is stronger); approximate values from the small Fig. 3 insets, rounded to the precision that can be read | Genotype group named first more bitter Tested, significant |
| The text says the geometric means in the insets of all 3 groups differed significantly from each other at every PTC concentration (p<0.001). Readings from a figure carry some margin. Individual lines spread widely within groups Where: Results, 'Suprathreshold Measures'; Fig. 3 insets For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same standard solution; numbers per group are not in the text, and volume and holding time are in a supplement that was not checked; values approximate from a figure | |||||
| Bitterness intensity (sensory) | PROP 550 µM, PAV/PAV vs AVI/AVI PROP 550 µM in water (the third concentration in Fig. 3) | about 12 | about 1–2 | bitterness intensity (group geometric mean of standardized gLMS ratings; higher is stronger); approximate values from the small Fig. 3 insets, rounded to the precision that can be read | Genotype group named first more bitter Tested, significant |
| The text says that for PROP, groups differed only at the second and third concentrations (170 and 550 µM in the text; 174 and 550 µM in the figure) (p<0.01), with PAV/PAV ratings significantly higher than the other 2 groups. Readings from a figure carry some margin. Where: Results, 'Suprathreshold Measures'; Fig. 3 insets For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same standard solution; numbers per group are not in the text, and volume and holding time are in a supplement that was not checked; values approximate from a figure | |||||
| Bitterness intensity (sensory) | PROP 550 µM, PAV/AVI vs AVI/AVI PROP 550 µM in water (the third concentration in Fig. 3) | about 3 | about 1–2 | bitterness intensity (group geometric mean of standardized gLMS ratings; higher is stronger); approximate values from the small Fig. 3 insets, rounded to the precision that can be read | Genotype group named first more bitter Test method described; result for this comparison not reported |
| The text says that for PROP, groups differed only at the second and third concentrations (170 and 550 µM in the text; 174 and 550 µM in the figure) (p<0.01), with PAV/PAV ratings significantly higher than the other 2 groups. No result is given for PAV/AVI vs AVI/AVI. Readings from a figure carry some margin. Where: Results, 'Suprathreshold Measures'; Fig. 3 insets For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same standard solution; numbers per group are not in the text, and volume and holding time are in a supplement that was not checked; values approximate from a figure | |||||
| Bitterness intensity (sensory) | PROP 550 µM, PAV/PAV vs PAV/AVI PROP 550 µM in water (the third concentration in Fig. 3) | about 12 | about 3 | bitterness intensity (group geometric mean of standardized gLMS ratings; higher is stronger); approximate values from the small Fig. 3 insets, rounded to the precision that can be read | Genotype group named first more bitter Tested, significant |
| The text says that for PROP, groups differed only at the second and third concentrations (170 and 550 µM in the text; 174 and 550 µM in the figure) (p<0.01), with PAV/PAV ratings significantly higher than the other 2 groups. Readings from a figure carry some margin. Where: Results, 'Suprathreshold Measures'; Fig. 3 insets For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same standard solution; numbers per group are not in the text, and volume and holding time are in a supplement that was not checked; values approximate from a figure | |||||
| Bitterness intensity (sensory) | PROP 5500 µM, PAV/PAV vs AVI/AVI PROP 5500 µM in water (the strongest solution in Fig. 3) | about 21 | about 12 | bitterness intensity (group geometric mean of standardized gLMS ratings; higher is stronger); approximate values from the small Fig. 3 insets, rounded to the precision that can be read | Genotype group named first more bitter Test method described; result for this comparison not reported |
| The text says group differences for PROP were reported only at the second and third concentrations; no test result for this pair at the strongest solution is given. The source notes that 2 AVI/AVI people responded strongly and 2 PAV/PAV people responded very little. Readings from a figure carry some margin. Where: Results, 'Suprathreshold Measures'; Fig. 3 insets For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same standard solution; numbers per group are not in the text, and volume and holding time are in a supplement that was not checked; values approximate from a figure | |||||
Duffy 2004 sensory only
Bitter Receptor Gene (TAS2R38), 6-n-Propylthiouracil (PROP) Bitterness and Alcohol Intake. Alcoholism: Clinical and Experimental Research 28(11):1629-1637 — doi:10.1097/01.alc.0000145789.55183.d4
- How deeply checked
- Checked in full text — Author manuscript on PMC (Methods; Results; Table 2); the Fig. 2 image was obtained from PMC and read by eye. The paper is mainly about alcohol; only the PROP ratings are used here, not the parts linked to drinking or health Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-24.
- Samples
- PROP in water; thresholds over 0.001–3.2 mM (quarter-log steps); suprathreshold ratings at 5 concentrations from 0.032 to 3.2 mM (half-log steps)
- Processing and cooking
- Not applicable (solutions only)
- How it was served or eaten
- Thresholds: two 10 mL samples (water and PROP), choosing the stronger, in an up-down procedure (geometric mean of reversals 2–7). Suprathreshold: 5 concentrations rated twice each
- Who rated, on what scale
- 84 healthy adults (53 women, 31 men; 36±13 years); genotypes PAV/PAV 21, PAV/AVI 37, AVI/AVI 26. Thresholds only for the first 48 (the F(2,46) in the text implies 49 people, so the numbers do not agree within the source). Intensity on the gLMS (0–100)
| Measured | Comparison and condition | Genotype group named first | Genotype group named second | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness threshold (sensory) | Threshold, PAV/PAV vs AVI/AVI PROP in water; up-down threshold (the first 48 participants) | 0.011 ± 0.003 mM | 0.579 ± 0.10 mM | threshold concentration (mM, mean ± SEM; lower means detected in a more dilute solution) | Genotype group named first detected bitterness at lower concentrations Tested; significance reported, but which pair differs is unclear |
| Thresholds differed by genotype (F(2,46)=89.783, p<0.0001). The text gives the means in the order AVI/AVI > PAV/AVI > PAV/PAV, but no pairwise test results. The 17 non-tasters defined by threshold and the 18 AVI/AVI people agreed closely, with 1 AVI/AVI person having a taster-range threshold (0.0468 mM) Where: Results, 'Relationship Between Genotype and Taste Phenotype'; Fig. 1 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups measured thresholds with the same solutions; numbers per group and pairwise tests are not given | |||||
| Bitterness threshold (sensory) | Threshold, PAV/AVI vs AVI/AVI PROP in water; up-down threshold (the first 48 participants) | 0.038 ± 0.001 mM | 0.579 ± 0.10 mM | threshold concentration (mM, mean ± SEM; lower means detected in a more dilute solution) | Genotype group named first detected bitterness at lower concentrations Tested; significance reported, but which pair differs is unclear |
| Thresholds differed by genotype (F(2,46)=89.783, p<0.0001). The text gives the means in the order AVI/AVI > PAV/AVI > PAV/PAV, but no pairwise test results. The 17 non-tasters defined by threshold and the 18 AVI/AVI people agreed closely, with 1 AVI/AVI person having a taster-range threshold (0.0468 mM) Where: Results, 'Relationship Between Genotype and Taste Phenotype'; Fig. 1 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups measured thresholds with the same solutions; numbers per group and pairwise tests are not given | |||||
| Bitterness threshold (sensory) | Threshold, PAV/PAV vs PAV/AVI PROP in water; up-down threshold (the first 48 participants) | 0.011 ± 0.003 mM | 0.038 ± 0.001 mM | threshold concentration (mM, mean ± SEM; lower means detected in a more dilute solution) | Genotype group named first detected bitterness at lower concentrations Tested; significance reported, but which pair differs is unclear |
| Thresholds differed by genotype (F(2,46)=89.783, p<0.0001). The text gives the means in the order AVI/AVI > PAV/AVI > PAV/PAV, but no pairwise test results. The 17 non-tasters defined by threshold and the 18 AVI/AVI people agreed closely, with 1 AVI/AVI person having a taster-range threshold (0.0468 mM) Where: Results, 'Relationship Between Genotype and Taste Phenotype'; Fig. 1 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups measured thresholds with the same solutions; numbers per group and pairwise tests are not given | |||||
| Bitterness intensity (sensory) | 3.2 mM, PAV/PAV vs AVI/AVI PROP 3.2 mM in water (the strongest of 5 concentrations) | about 50 | about 20 | bitterness intensity (gLMS 0–100, group mean; higher is stronger); approximate values from Fig. 2 | Genotype group named first more bitter Tested, significant |
| Genotype main effect F(2,81)=28.594, p<0.0001; in pairwise comparisons the 3 groups differed significantly from each other at the top 3 concentrations (PAV/PAV > PAV/AVI > AVI/AVI, p≤0.001). The text says PAV/PAV was "only slightly above" PAV/AVI. Readings from a figure carry some margin. When the same 84 people were grouped by their 3.2 mM ratings (bottom 25% non-tasters, top 25% supertasters, the rest medium), AVI/AVI (26) split into non-tasters 15, medium 9, supertasters 2; PAV/AVI (37) into 5, 23, 9; PAV/PAV (21) into 1, 10, 10 (Table 2; rank agreement Kendall τ=0.46). Pairs between rating-defined groups are not in this table Where: Results, 'Relationship Between Genotype and Taste Phenotype'; Fig. 2 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Usable — Genotype groups (84 people) rated the same solution on the gLMS, with pairwise tests; values approximate from a figure | |||||
| Bitterness intensity (sensory) | 3.2 mM, PAV/AVI vs AVI/AVI PROP 3.2 mM in water (the strongest of 5 concentrations) | about 40 | about 20 | bitterness intensity (gLMS 0–100, group mean; higher is stronger); approximate values from Fig. 2 | Genotype group named first more bitter Tested, significant |
| Genotype main effect F(2,81)=28.594, p<0.0001; in pairwise comparisons the 3 groups differed significantly from each other at the top 3 concentrations (PAV/PAV > PAV/AVI > AVI/AVI, p≤0.001). The text says PAV/PAV was "only slightly above" PAV/AVI. Readings from a figure carry some margin. When the same 84 people were grouped by their 3.2 mM ratings (bottom 25% non-tasters, top 25% supertasters, the rest medium), AVI/AVI (26) split into non-tasters 15, medium 9, supertasters 2; PAV/AVI (37) into 5, 23, 9; PAV/PAV (21) into 1, 10, 10 (Table 2; rank agreement Kendall τ=0.46). Pairs between rating-defined groups are not in this table Where: Results, 'Relationship Between Genotype and Taste Phenotype'; Fig. 2 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Usable — Genotype groups (84 people) rated the same solution on the gLMS, with pairwise tests; values approximate from a figure | |||||
| Bitterness intensity (sensory) | 3.2 mM, PAV/PAV vs PAV/AVI PROP 3.2 mM in water (the strongest of 5 concentrations) | about 50 | about 40 | bitterness intensity (gLMS 0–100, group mean; higher is stronger); approximate values from Fig. 2 | Genotype group named first more bitter Tested, significant |
| Genotype main effect F(2,81)=28.594, p<0.0001; in pairwise comparisons the 3 groups differed significantly from each other at the top 3 concentrations (PAV/PAV > PAV/AVI > AVI/AVI, p≤0.001). The text says PAV/PAV was "only slightly above" PAV/AVI. Readings from a figure carry some margin. When the same 84 people were grouped by their 3.2 mM ratings (bottom 25% non-tasters, top 25% supertasters, the rest medium), AVI/AVI (26) split into non-tasters 15, medium 9, supertasters 2; PAV/AVI (37) into 5, 23, 9; PAV/PAV (21) into 1, 10, 10 (Table 2; rank agreement Kendall τ=0.46). Pairs between rating-defined groups are not in this table Where: Results, 'Relationship Between Genotype and Taste Phenotype'; Fig. 2 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Usable — Genotype groups (84 people) rated the same solution on the gLMS, with pairwise tests; values approximate from a figure | |||||
Fischer 2014 sensory only
The Associations Between 6-N-propylthiouracil (PROP) Intensity and Taste Intensities Differ by TAS2R38 Haplotype. Journal of Nutrigenetics and Nutrigenomics 7(3):143-152 — doi:10.1159/000371552
- How deeply checked
- Checked in full text — Author manuscript on PMC (Methods; Results; Fig. 1); the issue is dated 2014 and online publication 2015 Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-24.
- Samples
- Filter-paper disks impregnated with 1.0 M PROP (not a solution held in the mouth; the concentration is that of the soaking solution); rated last, after disks for the other 4 tastes
- Processing and cooking
- Not applicable (solution soaked into paper)
- How it was served or eaten
- Not stated (how and how long the disk was placed is not in the text)
- Who rated, on what scale
- 1258 people aged 45 or over with only the common haplotypes (PAV/PAV 222, PAV/AVI 583, AVI/AVI 453); only those who completed gLMS (0–100) training. Group means are unadjusted; ANOVA with unadjusted pairwise comparisons
| Measured | Comparison and condition | Genotype group named first | Genotype group named second | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness intensity (sensory) | Paper disk, PAV/PAV vs AVI/AVI Filter-paper disk impregnated with 1.0 M PROP (concentration of the soaking solution) | 60.1 | 14.4 | bitterness intensity (gLMS 0–100, group mean; higher is stronger) | Genotype group named first more bitter Tested, significant |
| All pairwise comparisons p<0.0001 (ANOVA with unadjusted pairwise comparisons); genotype explained 35.6% of the variance in PROP intensity. Some PAV/PAV people rated very low and some AVI/AVI people very high. The values are not age- and sex-adjusted means Where: Methods, 'Statistical Analyses'; Results; Fig. 1 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Usable — Genotype groups (1258 people) rated the same disk on the gLMS, with pairwise tests; a filter-paper disk rather than a solution | |||||
| Bitterness intensity (sensory) | Paper disk, PAV/AVI vs AVI/AVI Filter-paper disk impregnated with 1.0 M PROP (concentration of the soaking solution) | 46.5 | 14.4 | bitterness intensity (gLMS 0–100, group mean; higher is stronger) | Genotype group named first more bitter Tested, significant |
| All pairwise comparisons p<0.0001 (ANOVA with unadjusted pairwise comparisons); genotype explained 35.6% of the variance in PROP intensity. Some PAV/PAV people rated very low and some AVI/AVI people very high. The values are not age- and sex-adjusted means Where: Methods, 'Statistical Analyses'; Results; Fig. 1 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Usable — Genotype groups (1258 people) rated the same disk on the gLMS, with pairwise tests; a filter-paper disk rather than a solution | |||||
| Bitterness intensity (sensory) | Paper disk, PAV/PAV vs PAV/AVI Filter-paper disk impregnated with 1.0 M PROP (concentration of the soaking solution) | 60.1 | 46.5 | bitterness intensity (gLMS 0–100, group mean; higher is stronger) | Genotype group named first more bitter Tested, significant |
| All pairwise comparisons p<0.0001 (ANOVA with unadjusted pairwise comparisons); genotype explained 35.6% of the variance in PROP intensity. Some PAV/PAV people rated very low and some AVI/AVI people very high. The values are not age- and sex-adjusted means Where: Methods, 'Statistical Analyses'; Results; Fig. 1 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Usable — Genotype groups (1258 people) rated the same disk on the gLMS, with pairwise tests; a filter-paper disk rather than a solution | |||||
Melis 2013 sensory only
The Gustin (CA6) Gene Polymorphism, rs2274333 (A/G), as a Mechanistic Link between PROP Tasting and Fungiform Taste Papilla Density and Maintenance. PLoS ONE 8(9):e74151 — doi:10.1371/journal.pone.0074151
- How deeply checked
- Checked in full text — Full text XML on PMC (Methods; Results; Fig. 1); Fig. 1 was checked as an image and values are approximate readings. The paper is mainly about the gustin gene and fungiform papillae; only the TAS2R38 group comparisons are used here Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-24.
- Samples
- PROP in water; thresholds over 0.00001–32 mM (quarter-log steps); intensity at 3.2 mM
- Processing and cooking
- Not applicable (solutions only)
- How it was served or eaten
- Intensity: 10 mL held for 10 s and spat out. Thresholds: ascending 3-AFC with 3 cups (on another day)
- Who rated, on what scale
- 63 young adults in Sardinia (text and table; the abstract says 64 and the sex breakdown also sums to 64, so the source is inconsistent); genotypes PAV/PAV 10, PAV/AVI 33, AVI/AVI 20. Intensity on the LMS; figures show mean±SEM; Newman–Keuls
| Measured | Comparison and condition | Genotype group named first | Genotype group named second | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness threshold (sensory) | Threshold, PAV/PAV vs AVI/AVI PROP in water; ascending 3-AFC threshold | about 0.2 mM | about 2.8 mM | threshold concentration (mM, group mean; lower means detected in a more dilute solution); approximate values from Fig. 1 | Genotype group named first detected bitterness at lower concentrations Tested, significant |
| Thresholds differed by TAS2R38 genotype (F[2,58]=6.0189, p=0.0042); AVI/AVI higher than the other 2 groups (p≤0.00158, Newman–Keuls). Readings from a figure carry some margin. Where: Results, 'PROP Thresholds and Bitterness Intensity'; Fig. 1 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups measured thresholds with the same solutions, with pairwise tests; values approximate from a figure; overlap of participants with other studies from the same laboratory cannot be checked | |||||
| Bitterness threshold (sensory) | Threshold, PAV/AVI vs AVI/AVI PROP in water; ascending 3-AFC threshold | about 0.9 mM | about 2.8 mM | threshold concentration (mM, group mean; lower means detected in a more dilute solution); approximate values from Fig. 1 | Genotype group named first detected bitterness at lower concentrations Tested, significant |
| Thresholds differed by TAS2R38 genotype (F[2,58]=6.0189, p=0.0042); AVI/AVI higher than the other 2 groups (p≤0.00158, Newman–Keuls). Readings from a figure carry some margin. Where: Results, 'PROP Thresholds and Bitterness Intensity'; Fig. 1 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups measured thresholds with the same solutions, with pairwise tests; values approximate from a figure; overlap of participants with other studies from the same laboratory cannot be checked | |||||
| Bitterness threshold (sensory) | Threshold, PAV/PAV vs PAV/AVI PROP in water; ascending 3-AFC threshold | about 0.2 mM | about 0.9 mM | threshold concentration (mM, group mean; lower means detected in a more dilute solution); approximate values from Fig. 1 | Genotype group named first detected bitterness at lower concentrations Tested, not significant |
| PAV/PAV vs PAV/AVI not significant (p>0.05), which does not mean the same. Readings from a figure carry some margin. Where: Results, 'PROP Thresholds and Bitterness Intensity'; Fig. 1 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups measured thresholds with the same solutions, with pairwise tests; values approximate from a figure; overlap of participants with other studies from the same laboratory cannot be checked | |||||
| Bitterness intensity (sensory) | 3.2 mM, PAV/PAV vs AVI/AVI 10 mL of 3.2 mM PROP in water for 10 s | about 77 | about 24.5 | bitterness intensity (LMS, group mean; higher is stronger); approximate values from Fig. 1 | Genotype group named first more bitter Tested, significant |
| Bitterness differed by TAS2R38 genotype (F[2,58]=32.468, p<0.000001); PAV/PAV > PAV/AVI (p≤0.0173), PAV/AVI > AVI/AVI (p=0.00011), and PAV/PAV vs AVI/AVI also significant by the different letters in the figure (Newman–Keuls). Readings from a figure carry some margin. Where: Results, 'PROP Thresholds and Bitterness Intensity'; Fig. 1 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same solution on the LMS, with pairwise tests; values approximate from a figure; overlap of participants with other studies from the same laboratory cannot be checked | |||||
| Bitterness intensity (sensory) | 3.2 mM, PAV/AVI vs AVI/AVI 10 mL of 3.2 mM PROP in water for 10 s | about 61.5 | about 24.5 | bitterness intensity (LMS, group mean; higher is stronger); approximate values from Fig. 1 | Genotype group named first more bitter Tested, significant |
| Bitterness differed by TAS2R38 genotype (F[2,58]=32.468, p<0.000001); PAV/PAV > PAV/AVI (p≤0.0173), PAV/AVI > AVI/AVI (p=0.00011), and PAV/PAV vs AVI/AVI also significant by the different letters in the figure (Newman–Keuls). Readings from a figure carry some margin. Where: Results, 'PROP Thresholds and Bitterness Intensity'; Fig. 1 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same solution on the LMS, with pairwise tests; values approximate from a figure; overlap of participants with other studies from the same laboratory cannot be checked | |||||
| Bitterness intensity (sensory) | 3.2 mM, PAV/PAV vs PAV/AVI 10 mL of 3.2 mM PROP in water for 10 s | about 77 | about 61.5 | bitterness intensity (LMS, group mean; higher is stronger); approximate values from Fig. 1 | Genotype group named first more bitter Tested, significant |
| Bitterness differed by TAS2R38 genotype (F[2,58]=32.468, p<0.000001); PAV/PAV > PAV/AVI (p≤0.0173), PAV/AVI > AVI/AVI (p=0.00011), and PAV/PAV vs AVI/AVI also significant by the different letters in the figure (Newman–Keuls). Readings from a figure carry some margin. Where: Results, 'PROP Thresholds and Bitterness Intensity'; Fig. 1 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same solution on the LMS, with pairwise tests; values approximate from a figure; overlap of participants with other studies from the same laboratory cannot be checked | |||||
Melis 2015 sensory only
Dose-Dependent Effects of L-Arginine on PROP Bitterness Intensity and Latency and Characteristics of the Chemical Interaction between PROP and L-Arginine. PLoS ONE 10(6):e0131104 — doi:10.1371/journal.pone.0131104
- How deeply checked
- Checked in full text — Full text XML on PMC (Methods; Results; Fig. 4); Fig. 4 was checked as an image and values are approximate readings. The paper is mainly about PROP with added L-arginine; only group comparisons for plain PROP solution are used here Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-24.
- Samples
- 3.2 mM PROP in water (nothing added)
- Processing and cooking
- Not applicable (solutions only)
- How it was served or eaten
- 4 mL swished in the mouth until a taste was felt (up to about 2 min), then spat out
- Who rated, on what scale
- 51 people aged 20–35 (14 men, 37 women); genotypes PAV/PAV 10, PAV/AVI 23, AVI/AVI 18. Intensity on the LMS; figures show mean±SEM; Newman–Keuls. Taster-status groups based on PROP ratings also exist but are not in this table
| Measured | Comparison and condition | Genotype group named first | Genotype group named second | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness intensity (sensory) | 3.2 mM, PAV/PAV vs AVI/AVI 4 mL of 3.2 mM PROP in water (nothing added) | about 69 | about 3 | bitterness intensity (LMS 0–100, group mean; higher is stronger); approximate values from Fig. 4A | Genotype group named first more bitter Tested, significant |
| Bitterness differed by genotype (F[2,48]>31.392, p<0.00001); AVI/AVI lower than the other 2 groups (p=0.000128, Newman–Keuls). The 8 people who felt no bitterness from the plain 3.2 mM solution were all AVI/AVI. Readings from a figure carry some margin. Where: Results following "PROP taster status and genotyping"; Fig. 4A For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same solution on the LMS, with pairwise tests; values approximate from a figure; overlap of participants with other studies from the same laboratory cannot be checked | |||||
| Bitterness intensity (sensory) | 3.2 mM, PAV/AVI vs AVI/AVI 4 mL of 3.2 mM PROP in water (nothing added) | about 51 | about 3 | bitterness intensity (LMS 0–100, group mean; higher is stronger); approximate values from Fig. 4A | Genotype group named first more bitter Tested, significant |
| Bitterness differed by genotype (F[2,48]>31.392, p<0.00001); AVI/AVI lower than the other 2 groups (p=0.000128, Newman–Keuls). The 8 people who felt no bitterness from the plain 3.2 mM solution were all AVI/AVI. Readings from a figure carry some margin. Where: Results following "PROP taster status and genotyping"; Fig. 4A For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same solution on the LMS, with pairwise tests; values approximate from a figure; overlap of participants with other studies from the same laboratory cannot be checked | |||||
| Bitterness intensity (sensory) | 3.2 mM, PAV/PAV vs PAV/AVI 4 mL of 3.2 mM PROP in water (nothing added) | about 69 | about 51 | bitterness intensity (LMS 0–100, group mean; higher is stronger); approximate values from Fig. 4A | Genotype group named first more bitter Tested, not significant |
| PAV/PAV vs PAV/AVI not significant (p=0.12), which does not mean the same. Readings from a figure carry some margin. Where: Results following "PROP taster status and genotyping"; Fig. 4A For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same solution on the LMS, with pairwise tests; values approximate from a figure; overlap of participants with other studies from the same laboratory cannot be checked | |||||
Sollai 2017 sensory only
First objective evaluation of taste sensitivity to 6-n-propylthiouracil (PROP), a paradigm gustatory stimulus in humans. Scientific Reports 7:40353 — doi:10.1038/srep40353
- How deeply checked
- Checked in full text — Full text XML on PMC (Methods; Results; Fig. 1); Fig. 1 was checked as an image and medians are approximate readings. The paper is mainly about tongue electrophysiology; only group comparisons of bitterness ratings are used here Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-24.
- Samples
- A 6 mm paper disk impregnated with 30 µL of 3.2 mM PROP solution (not a solution held in the mouth)
- Processing and cooking
- Not applicable (solution soaked into paper)
- How it was served or eaten
- The disk was placed on a small circular area at the tongue tip for 15 s (during electrical recording)
- Who rated, on what scale
- 43 of 45 tested, after excluding 2 whose taster classification differed between two methods (the 3-solution test and a 50 mM impregnated paper); genotypes PAV/PAV 7, PAV/AVI 25, AVI/AVI 11. Intensity on the LMS; box plots (medians); Newman–Keuls
| Measured | Comparison and condition | Genotype group named first | Genotype group named second | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness intensity (sensory) | Paper disk, PAV/PAV vs AVI/AVI Paper disk with 30 µL of 3.2 mM PROP on the tongue tip for 15 s | about 60 | about 3 | bitterness intensity (LMS, group median; higher is stronger); approximate values from Fig. 1 | Genotype group named first more bitter Tested, significant |
| Bitterness differed by genotype (F 2,40 =11.294, P=0.00014); AVI/AVI lower than the other 2 groups (P<0.0016, Newman–Keuls). Readings from a figure carry some margin. Where: Start of Results; Fig. 1; Methods, 'PROP taste stimulations' For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same paper disk on the LMS, with pairwise tests; values are approximate medians from a figure, rated during electrical recording; overlap of participants with other studies from the same laboratory cannot be checked | |||||
| Bitterness intensity (sensory) | Paper disk, PAV/AVI vs AVI/AVI Paper disk with 30 µL of 3.2 mM PROP on the tongue tip for 15 s | about 35 | about 3 | bitterness intensity (LMS, group median; higher is stronger); approximate values from Fig. 1 | Genotype group named first more bitter Tested, significant |
| Bitterness differed by genotype (F 2,40 =11.294, P=0.00014); AVI/AVI lower than the other 2 groups (P<0.0016, Newman–Keuls). Readings from a figure carry some margin. Where: Start of Results; Fig. 1; Methods, 'PROP taste stimulations' For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same paper disk on the LMS, with pairwise tests; values are approximate medians from a figure, rated during electrical recording; overlap of participants with other studies from the same laboratory cannot be checked | |||||
| Bitterness intensity (sensory) | Paper disk, PAV/PAV vs PAV/AVI Paper disk with 30 µL of 3.2 mM PROP on the tongue tip for 15 s | about 60 | about 35 | bitterness intensity (LMS, group median; higher is stronger); approximate values from Fig. 1 | Genotype group named first more bitter Tested, not significant |
| PAV/PAV vs PAV/AVI not significant (P>0.05), which does not mean the same. Readings from a figure carry some margin. Where: Start of Results; Fig. 1; Methods, 'PROP taste stimulations' For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same paper disk on the LMS, with pairwise tests; values are approximate medians from a figure, rated during electrical recording; overlap of participants with other studies from the same laboratory cannot be checked | |||||
Nolden 2024 sensory only
Differential Activation of TAS2R4 May Recover Ability to Taste Propylthiouracil for Some TAS2R38 AVI Homozygotes. Nutrients 16(9):1357 — doi:10.3390/nu16091357
- How deeply checked
- Checked in full text — Full text XML on PMC (Methods; Results 3.3; Table 1; Fig. 3). Fig. 3 has no numeric axis, so values are approximations converted from the positions of the verbal labels (using the gLMS label positions stated in the 2020 paper from the same study). Values from Nolden 2020, on the first-session participants of the same study, are not moved into this row Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-24.
- Samples
- PROP in water at 0.032, 0.1, 0.32, 1.0 and 3.2 mM (half-log steps)
- Processing and cooking
- Not applicable (solutions only)
- How it was served or eaten
- Each concentration rated twice (rinsing with water between samples)
- Who rated, on what scale
- 218 of 243 people (mean 25.8 years) with the 3 main genotypes (PAV/PAV 48, PAV/AVI 119, AVI/AVI 51); gLMS; Tukey–Kramer
| Measured | Comparison and condition | Genotype group named first | Genotype group named second | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness intensity (sensory) | 3.2 mM, PAV/PAV vs AVI/AVI PROP 3.2 mM in water (the strongest of 5 concentrations) | about 55 | about 20 | bitterness intensity (gLMS, group mean; higher is stronger); approximate values from Fig. 3, converted using the scale positions stated in the 2020 paper from the same study | Genotype group named first more bitter Tested, significant |
| TAS2R38 diplotype explained 29.3% of the variance in bitterness at 3.2 mM (F(1,216)=89.4, p<0.0001); the text says the 3 groups differed significantly from each other at the top 3 concentrations (Tukey–Kramer; no pairwise p values given). The figure has no numeric axis; values are approximations converted from the positions of the verbal labels (moderate 17, strong 35, very strong 51, as stated in the 2020 paper from the same study). The means in the 2020 paper (the 220 first-session participants), judged from its procedure and numbers to belong to the same series of experiments, are not moved into this row. The paper focuses on some AVI/AVI people perceiving strong bitterness Where: Results 3.3; Fig. 3; Table 1 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups (218 people) rated the same solution on the gLMS, with pairwise comparisons; values are approximations from a figure converted from verbal labels | |||||
| Bitterness intensity (sensory) | 3.2 mM, PAV/AVI vs AVI/AVI PROP 3.2 mM in water (the strongest of 5 concentrations) | about 44 | about 20 | bitterness intensity (gLMS, group mean; higher is stronger); approximate values from Fig. 3, converted using the scale positions stated in the 2020 paper from the same study | Genotype group named first more bitter Tested, significant |
| TAS2R38 diplotype explained 29.3% of the variance in bitterness at 3.2 mM (F(1,216)=89.4, p<0.0001); the text says the 3 groups differed significantly from each other at the top 3 concentrations (Tukey–Kramer; no pairwise p values given). The figure has no numeric axis; values are approximations converted from the positions of the verbal labels (moderate 17, strong 35, very strong 51, as stated in the 2020 paper from the same study). The means in the 2020 paper (the 220 first-session participants), judged from its procedure and numbers to belong to the same series of experiments, are not moved into this row. The paper focuses on some AVI/AVI people perceiving strong bitterness Where: Results 3.3; Fig. 3; Table 1 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups (218 people) rated the same solution on the gLMS, with pairwise comparisons; values are approximations from a figure converted from verbal labels | |||||
| Bitterness intensity (sensory) | 3.2 mM, PAV/PAV vs PAV/AVI PROP 3.2 mM in water (the strongest of 5 concentrations) | about 55 | about 44 | bitterness intensity (gLMS, group mean; higher is stronger); approximate values from Fig. 3, converted using the scale positions stated in the 2020 paper from the same study | Genotype group named first more bitter Tested, significant |
| TAS2R38 diplotype explained 29.3% of the variance in bitterness at 3.2 mM (F(1,216)=89.4, p<0.0001); the text says the 3 groups differed significantly from each other at the top 3 concentrations (Tukey–Kramer; no pairwise p values given). The figure has no numeric axis; values are approximations converted from the positions of the verbal labels (moderate 17, strong 35, very strong 51, as stated in the 2020 paper from the same study). The means in the 2020 paper (the 220 first-session participants), judged from its procedure and numbers to belong to the same series of experiments, are not moved into this row. The paper focuses on some AVI/AVI people perceiving strong bitterness Where: Results 3.3; Fig. 3; Table 1 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups (218 people) rated the same solution on the gLMS, with pairwise comparisons; values are approximations from a figure converted from verbal labels | |||||
Aoki 2023 sensory only
Association between Genetic Variation in the TAS2R38 Bitter Taste Receptor and Propylthiouracil Bitter Taste Thresholds among Adults Living in Japan Using the Modified 2AFC Procedure with the Quest Method. Nutrients 15(10):2415 — doi:10.3390/nu15102415
- How deeply checked
- Checked in full text — Full text XML on PMC (Methods 2.3; Results 3.2; Table 2; Fig. 4); Fig. 4 was checked as an image and medians are approximate readings (log axis) Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-24.
- Samples
- PROP in water, 20 steps from 0.002 to 3.2 mM (one-sixth-log steps)
- Processing and cooking
- Not applicable (solutions only)
- How it was served or eaten
- 10 mL of water and of PROP taken in turn and spat out, answering which had a taste (modified 2AFC); the 75%-correct point estimated by QUEST was the detection threshold. Steps up in concentration were limited to avoid a sudden strong bitterness
- Who rated, on what scale
- Adults living in Japan; 118 genotyped, thresholds measured in 79 (PAV/PAV 24, PAV/AVI 43, AVI/AVI 12). ANOVA on log thresholds; Tukey
| Measured | Comparison and condition | Genotype group named first | Genotype group named second | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness threshold (sensory) | Threshold, PAV/PAV vs AVI/AVI 10 mL of PROP in water; detection threshold by modified 2AFC with QUEST | about 0.025 mM | about 1.3 mM | detection threshold concentration (mM, group median; lower means detected in a more dilute solution); approximate values from Fig. 4 (log axis) | Genotype group named first detected bitterness at lower concentrations Tested, significant |
| Genotype effect on log thresholds (F(2,75)=161.46, p<0.001, with age as covariate); by Tukey, p<0.001. Readings from a figure carry some margin. Where: Methods 2.3; Results 3.2; Table 2; Fig. 4 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups (79 people) measured thresholds by the same procedure, with pairwise tests; values are approximate medians from a figure; AVI/AVI had 12 people | |||||
| Bitterness threshold (sensory) | Threshold, PAV/AVI vs AVI/AVI 10 mL of PROP in water; detection threshold by modified 2AFC with QUEST | about 0.043 mM | about 1.3 mM | detection threshold concentration (mM, group median; lower means detected in a more dilute solution); approximate values from Fig. 4 (log axis) | Genotype group named first detected bitterness at lower concentrations Tested, significant |
| Genotype effect on log thresholds (F(2,75)=161.46, p<0.001, with age as covariate); by Tukey, p<0.001. Readings from a figure carry some margin. Where: Methods 2.3; Results 3.2; Table 2; Fig. 4 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups (79 people) measured thresholds by the same procedure, with pairwise tests; values are approximate medians from a figure; AVI/AVI had 12 people | |||||
| Bitterness threshold (sensory) | Threshold, PAV/PAV vs PAV/AVI 10 mL of PROP in water; detection threshold by modified 2AFC with QUEST | about 0.025 mM | about 0.043 mM | detection threshold concentration (mM, group median; lower means detected in a more dilute solution); approximate values from Fig. 4 (log axis) | Genotype group named first detected bitterness at lower concentrations Tested, significant |
| Genotype effect on log thresholds (F(2,75)=161.46, p<0.001, with age as covariate); by Tukey, p<0.01; still significant at p<0.05 after removing the lowest PAV/PAV value. Readings from a figure carry some margin. Where: Methods 2.3; Results 3.2; Table 2; Fig. 4 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups (79 people) measured thresholds by the same procedure, with pairwise tests; values are approximate medians from a figure; AVI/AVI had 12 people | |||||
Campbell 2012 sensory only
Evolution of Functionally Diverse Alleles Associated with PTC Bitter Taste Sensitivity in Africa. Molecular Biology and Evolution 29(4):1141-1153 — doi:10.1093/molbev/msr293
- How deeply checked
- Checked in full text — Full text on PMC (Methods; Results); Fig. 3 and the supplement (age and sex adjustment, test tables) were not obtained. The paper is mainly about the evolution of the gene Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-24.
- Samples
- Vials of a PTC dilution series (13, 11, 9, 7, 5, 3, 1; 13 is the most dilute)
- Processing and cooking
- Not applicable (solutions only)
- How it was served or eaten
- In the field, with a modified Harris–Kalmus recognition method, recording the most dilute number at which bitterness was recognised
- Who rated, on what scale
- 463 people in Cameroon and Kenya; numbers per group not stated. Values are unadjusted scores (whether the ± in the text is SD or SE is not stated)
| Measured | Comparison and condition | Genotype group named first | Genotype group named second | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness threshold (sensory) | PTC dilution score, PAV/PAV vs AVI/AVI PTC dilution series (13 most dilute, 1 most concentrated); recognition method in the field | 9.51 ± 1.66 | 3.36 ± 2.11 | number of the most dilute solution at which bitterness was recognised (group mean; higher means recognised in a more dilute solution) | Genotype group named first detected bitterness at lower concentrations Tested; significance reported, but which pair differs is unclear |
| Linear regression and Kruskal–Wallis showed an association with genotype (both P<0.001); these are overall tests across many genotypes, not tests of this pair. Whether pairwise tests exist in the supplement (not obtained) cannot be confirmed. Values are scores not adjusted for age and sex; whether ± is SD or SE is not stated. Rare combinations (AAI/AVI and others) were intermediate (6.32 ± 2.06) Where: Results, 'Association between genotype and phenotype' For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups were tested with the same dilution series, but the test result for this pair cannot be confirmed in the text obtained and checked, and numbers per group are not given; an African sample with many rare haplotypes, differing in ancestry and method from the other studies | |||||
Graham 2021 sensory only
The Associations Between Bitter and Fat Taste Sensitivity, and Dietary Fat Intake: Are They Impacted by Genetic Predisposition?. Chemical Senses 46:bjab029 — doi:10.1093/chemse/bjab029
- How deeply checked
- Checked in full text — Full text on PMC (Methods; Results; Fig. 1 legend); medians and interquartile ranges are from the text. The paper is mainly about fat taste and diet; only group comparisons of PTC ratings are used here Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-24.
- Samples
- A commercial PTC-impregnated strip (EISCO FSC1031; amount and concentration not stated)
- Processing and cooking
- Not applicable (solution soaked into paper)
- How it was served or eaten
- Not stated (how and how long the strip was applied is not in the text)
- Who rated, on what scale
- 88 white adults in the UK; PAV/PAV 13, AVI/PAV 41, AVI/AVI 24, PAV/AAV 3, AVI/AAV 7; gLMS (0–100); Kruskal–Wallis with Bonferroni-adjusted pairwise comparisons
| Measured | Comparison and condition | Genotype group named first | Genotype group named second | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness intensity (sensory) | Strip, PAV/PAV vs AVI/AVI Commercial PTC-impregnated strip (amount and concentration not stated) | 31 [30] | 2 [6] | bitterness intensity (gLMS 0–100, group median [interquartile range]; higher is stronger) | Genotype group named first more bitter Tested, significant |
| Kruskal–Wallis with Bonferroni-adjusted pairwise comparison: P=1×10⁻⁶ Where: Results, 'Genetics and taste sensitivity'; Fig. 1A legend For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same commercial strip on the gLMS, with tests for some pairs; the strip concentration is unknown and PAV/PAV had 13 people | |||||
| Bitterness intensity (sensory) | Strip, PAV/AVI vs AVI/AVI Commercial PTC-impregnated strip (amount and concentration not stated) | 9 [24] | 2 [6] | bitterness intensity (gLMS 0–100, group median [interquartile range]; higher is stronger) | Genotype group named first more bitter Tested, significant |
| Kruskal–Wallis with Bonferroni-adjusted pairwise comparison: P=0.002 (written AVI/PAV in the source) Where: Results, 'Genetics and taste sensitivity'; Fig. 1A legend For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same commercial strip on the gLMS, with tests for some pairs; the strip concentration is unknown and PAV/PAV had 13 people | |||||
| Bitterness intensity (sensory) | Strip, PAV/PAV vs PAV/AVI Commercial PTC-impregnated strip (amount and concentration not stated) | 31 [30] | 9 [24] | bitterness intensity (gLMS 0–100, group median [interquartile range]; higher is stronger) | Genotype group named first more bitter Test method described; result for this comparison not reported |
| The test method (Kruskal–Wallis with Bonferroni pairwise comparisons) is described, but no result is given for this pair Where: Results, 'Genetics and taste sensitivity'; Fig. 1A legend For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups rated the same commercial strip on the gLMS, with tests for some pairs; the strip concentration is unknown and PAV/PAV had 13 people | |||||
Smutzer 2013 sensory only
Validation of Edible Taste Strips for Assessing PROP Taste Perception. Chemical Senses 38(6):529-539 — doi:10.1093/chemse/bjt023
- How deeply checked
- Checked in full text — Full text on PMC (Methods; Results); the Fig. 1 image was obtained from PMC and read by eye (the axis is broken above 60) Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-24.
- Samples
- PROP in water at 0.032, 0.32 and 3.2 mM, and dissolvable films (1-inch squares made of pullulan and other polymers) containing 400 or 600 nmol PROP (amount per film)
- Processing and cooking
- Films were made by pouring and drying a polymer mixture containing PROP; not applicable for solutions
- How it was served or eaten
- Films rated first, then solutions; a film was placed on the tongue, pressed to the roof of the mouth and swallowed; solutions 10 mL; rinsing twice with water before and after
- Who rated, on what scale
- 108 of 118 after excluding 1 outlier and 9 with rare haplotypes (PAV/PAV 35, PAV/AVI 38, AVI/AVI 35); genotyped at the Monell Chemical Senses Center; gLMS; mean±SEM; Newman–Keuls
| Measured | Comparison and condition | Genotype group named first | Genotype group named second | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness intensity (sensory) | Solution 3.2 mM, PAV/PAV vs AVI/AVI 10 mL of 3.2 mM PROP in water | about 61.5 | about 15 | bitterness intensity (gLMS, group mean; higher is stronger); approximate values from Fig. 1B | Genotype group named first more bitter Tested, significant |
| Genotype effect at 3.2 mM (F(2,105)=48.5); the text says all Newman–Keuls pairwise comparisons were P<0.05 except PAV/PAV vs AVI/AVI at 0.032 mM (P=0.09) and PAV/PAV vs PAV/AVI at 0.32 mM (P=0.06); at 3.2 mM this covers all 3 pairs of groups. Readings from a figure carry some margin. Where: Results, 'Taste intensity values for PROP strips and PROP solutions'; Fig. 1 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Usable — Genotype groups (108 people) rated the same solution on the gLMS, with pairwise tests; values approximate from a figure | |||||
| Bitterness intensity (sensory) | Solution 3.2 mM, PAV/AVI vs AVI/AVI 10 mL of 3.2 mM PROP in water | about 50.5 | about 15 | bitterness intensity (gLMS, group mean; higher is stronger); approximate values from Fig. 1B | Genotype group named first more bitter Tested, significant |
| Genotype effect at 3.2 mM (F(2,105)=48.5); the text says all Newman–Keuls pairwise comparisons were P<0.05 except PAV/PAV vs AVI/AVI at 0.032 mM (P=0.09) and PAV/PAV vs PAV/AVI at 0.32 mM (P=0.06); at 3.2 mM this covers all 3 pairs of groups. Readings from a figure carry some margin. Where: Results, 'Taste intensity values for PROP strips and PROP solutions'; Fig. 1 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Usable — Genotype groups (108 people) rated the same solution on the gLMS, with pairwise tests; values approximate from a figure | |||||
| Bitterness intensity (sensory) | Solution 3.2 mM, PAV/PAV vs PAV/AVI 10 mL of 3.2 mM PROP in water | about 61.5 | about 50.5 | bitterness intensity (gLMS, group mean; higher is stronger); approximate values from Fig. 1B | Genotype group named first more bitter Tested, significant |
| Genotype effect at 3.2 mM (F(2,105)=48.5); the text says all Newman–Keuls pairwise comparisons were P<0.05 except PAV/PAV vs AVI/AVI at 0.032 mM (P=0.09) and PAV/PAV vs PAV/AVI at 0.32 mM (P=0.06); at 3.2 mM this covers all 3 pairs of groups. Readings from a figure carry some margin. Where: Results, 'Taste intensity values for PROP strips and PROP solutions'; Fig. 1 For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Usable — Genotype groups (108 people) rated the same solution on the gLMS, with pairwise tests; values approximate from a figure | |||||
| Bitterness intensity (sensory) | Film 600 nmol, PAV/PAV vs AVI/AVI Dissolvable film containing 600 nmol PROP (amount in 1 film, not a solution concentration) | about 43.5 | about 13 | bitterness intensity (gLMS, group mean; higher is stronger); approximate values from Fig. 1A | Genotype group named first more bitter Tested, significant |
| Genotype effect for films (600 nmol: F(2,105)=19.8, P<0.0001); PAV/PAV and PAV/AVI higher than AVI/AVI (Newman–Keuls, P<0.001). Readings from a figure carry some margin. Where: Results, 'Taste intensity values for PROP strips and PROP solutions'; Fig. 1; Discussion For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups (108 people) rated the same film on the gLMS, with tests; a film rather than a solution, with the amount per film; values approximate from a figure | |||||
| Bitterness intensity (sensory) | Film 600 nmol, PAV/AVI vs AVI/AVI Dissolvable film containing 600 nmol PROP (amount in 1 film, not a solution concentration) | about 37 | about 13 | bitterness intensity (gLMS, group mean; higher is stronger); approximate values from Fig. 1A | Genotype group named first more bitter Tested, significant |
| Genotype effect for films (600 nmol: F(2,105)=19.8, P<0.0001); PAV/PAV and PAV/AVI higher than AVI/AVI (Newman–Keuls, P<0.001). Readings from a figure carry some margin. Where: Results, 'Taste intensity values for PROP strips and PROP solutions'; Fig. 1; Discussion For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups (108 people) rated the same film on the gLMS, with tests; a film rather than a solution, with the amount per film; values approximate from a figure | |||||
| Bitterness intensity (sensory) | Film 600 nmol, PAV/PAV vs PAV/AVI Dissolvable film containing 600 nmol PROP (amount in 1 film, not a solution concentration) | about 43.5 | about 37 | bitterness intensity (gLMS, group mean; higher is stronger); approximate values from Fig. 1A | Genotype group named first more bitter Tested, not significant |
| The discussion says there were no statistical differences between PAV/PAV and PAV/AVI for films (no p value given), which does not mean the same; the 400 nmol film shows the same pattern. Readings from a figure carry some margin. Where: Results, 'Taste intensity values for PROP strips and PROP solutions'; Fig. 1; Discussion For whether, when groups of people with different TAS2R38 genotypes rate the same standard PROP or PTC stimulus (solution, paper soaked in solution, or dissolvable film), bitterness intensity or thresholds differ between the groups: Limited — Genotype groups (108 people) rated the same film on the gLMS, with tests; a film rather than a solution, with the amount per film; values approximate from a figure | |||||
What was established within an experiment
Only results comparing genotype groups that rated the same stimulus within one study. Results without a test are stated as such.
- In Duffy and colleagues, the same 84 people rated a 3.2 mM PROP solution: about 50 for PAV/PAV, about 40 for PAV/AVI and about 20 for AVI/AVI (approximate figure readings), and the 3 groups differed significantly from each other. The authors say PAV/PAV was "only slightly above" PAV/AVI. [rows: Duffy 2004 3.2 mM, PAV/PAV vs AVI/AVI; Duffy 2004 3.2 mM, PAV/AVI vs AVI/AVI; Duffy 2004 3.2 mM, PAV/PAV vs PAV/AVI]
- In Fischer and colleagues, 1258 people aged 45 or over rated a filter-paper disk impregnated with 1.0 M PROP; the means were 60.1 for PAV/PAV, 46.5 for PAV/AVI and 14.4 for AVI/AVI, with p<0.0001 for every pair. [rows: Fischer 2014 Paper disk, PAV/PAV vs AVI/AVI; Fischer 2014 Paper disk, PAV/AVI vs AVI/AVI; Fischer 2014 Paper disk, PAV/PAV vs PAV/AVI]
- The direction was the same for thresholds. Among 79 adults in Japan in Aoki and colleagues, median detection thresholds were about 0.025 mM for PAV/PAV, about 0.043 mM for PAV/AVI and about 1.3 mM for AVI/AVI (approximate figure readings), with significant differences in all 3 pairs. Bufe and colleagues found PTC recognition thresholds of 25.0, 40.6 and 4730 µM, and the authors write that PAV/PAV and PAV/AVI did not differ significantly. [rows: Aoki 2023 Threshold, PAV/PAV vs AVI/AVI; Aoki 2023 Threshold, PAV/AVI vs AVI/AVI; Aoki 2023 Threshold, PAV/PAV vs PAV/AVI; Bufe 2005 PTC recognition threshold, PAV/PAV vs AVI/AVI; Bufe 2005 PTC recognition threshold, PAV/AVI vs AVI/AVI; Bufe 2005 PTC recognition threshold, PAV/PAV vs PAV/AVI]
- In Bufe and colleagues, genotype differences behaved differently for PTC and PROP. At 5600 µM PTC the 3 groups differed significantly from each other, but for PROP differences appeared only at the 2 middle concentrations, and no significant difference was shown at the strongest, 5500 µM. For PROP, the source notes that 2 AVI/AVI people responded strongly and 2 PAV/PAV people responded very little. [rows: Bufe 2005 PTC 5600 µM, PAV/PAV vs AVI/AVI; Bufe 2005 PTC 5600 µM, PAV/AVI vs AVI/AVI; Bufe 2005 PTC 5600 µM, PAV/PAV vs PAV/AVI; Bufe 2005 PROP 550 µM, PAV/PAV vs AVI/AVI; Bufe 2005 PROP 5500 µM, PAV/PAV vs AVI/AVI]
- In Smutzer and colleagues, the same 108 people rated both solutions and films. For the 3.2 mM solution the 3 groups differed significantly from each other, but for the 600 nmol film the authors write there was no statistical difference between PAV/PAV and PAV/AVI. [rows: Smutzer 2013 Solution 3.2 mM, PAV/PAV vs AVI/AVI; Smutzer 2013 Solution 3.2 mM, PAV/AVI vs AVI/AVI; Smutzer 2013 Solution 3.2 mM, PAV/PAV vs PAV/AVI; Smutzer 2013 Film 600 nmol, PAV/PAV vs AVI/AVI; Smutzer 2013 Film 600 nmol, PAV/AVI vs AVI/AVI; Smutzer 2013 Film 600 nmol, PAV/PAV vs PAV/AVI]
Candidate explanations for the differences
Conditions that differ between studies and might explain differing results, and the authors' interpretations. Several conditions differ at once, so which one mattered is unknown. These are candidates, not causes.
- One candidate explanation for why the PAV/PAV vs PAV/AVI difference splits across studies is differences in presentation and concentration. Even in the same participants in Smutzer and colleagues, the solution showed a significant difference between PAV/PAV and PAV/AVI and the film did not. Solutions, filter-paper disks, strips and films measure the amount entering the mouth differently, and the concentration for paper is that of the soaking solution, not the concentration in the mouth. [rows: Smutzer 2013 Solution 3.2 mM, PAV/PAV vs PAV/AVI; Smutzer 2013 Film 600 nmol, PAV/PAV vs PAV/AVI; Fischer 2014 Paper disk, PAV/PAV vs PAV/AVI; Sollai 2017 Paper disk, PAV/PAV vs PAV/AVI]
- Thresholds and intensity ratings of strong solutions are different measures, and genotype may act differently on them. In the same participants of Melis and colleagues in 2013, thresholds did not differ significantly between PAV/PAV and PAV/AVI, while intensity at 3.2 mM did. Bufe and colleagues also write that PAV/AVI recognition thresholds were close to PAV/PAV while suprathreshold intensity was intermediate. [rows: Melis 2013 Threshold, PAV/PAV vs PAV/AVI; Melis 2013 3.2 mM, PAV/PAV vs PAV/AVI; Bufe 2005 PTC recognition threshold, PAV/PAV vs PAV/AVI; Bufe 2005 PTC 5600 µM, PAV/PAV vs PAV/AVI]
- Group size is another candidate. Melis and colleagues in 2015, where no significant difference was shown, had 10 PAV/PAV people and Sollai and colleagues had 7, while Fischer and colleagues had 222 and 583. With small groups, a difference is harder to detect even if it exists. [rows: Melis 2015 3.2 mM, PAV/PAV vs PAV/AVI; Sollai 2017 Paper disk, PAV/PAV vs PAV/AVI; Fischer 2014 Paper disk, PAV/PAV vs PAV/AVI]
- Age and population are further candidates. In Mennella and colleagues, age differences appeared only within PAV/AVI (described in the scope table). The African sample of Campbell and colleagues had many rare haplotypes, with combinations of intermediate sensitivity. In Aoki and colleagues in Japan the share of PAV/PAV was high. The authors suggest that other genes (TAS2R4, gustin) or saliva components may be involved, but this cannot be confirmed from the comparisons in this table. [rows: Campbell 2012 PTC dilution score, PAV/PAV vs AVI/AVI; Aoki 2023 Threshold, PAV/PAV vs AVI/AVI; Nolden 2024 3.2 mM, PAV/PAV vs AVI/AVI; Melis 2013 3.2 mM, PAV/PAV vs AVI/AVI]
What is not known — what this comparison alone cannot separate
- Groups defined by PROP ratings, such as "supertasters", are not used as pairs in this table. When groups are defined by the same 3.2 mM ratings that are then compared, as in Duffy and colleagues, the difference largely follows from the definition. When PROP was rated on another day or with another presentation than the classification test, as in Melis and colleagues in 2015 or Sollai and colleagues, the result is not fixed by the definition alone, but because the classification uses PROP ratings it is not treated as evidence independent of genotype. In Table 2 of Duffy and colleagues, when the same 84 people were grouped by 3.2 mM ratings, there were 2 supertasters among the 26 AVI/AVI people and 1 non-taster among the 21 PAV/PAV people: genotype and rating-based classes overlapped but did not match. [rows: Duffy 2004 3.2 mM, PAV/PAV vs AVI/AVI; Melis 2015 3.2 mM, PAV/PAV vs AVI/AVI; Sollai 2017 Paper disk, PAV/PAV vs AVI/AVI]
- The number of rows is not the number of independent studies. The three pair rows of a study share participants, and whether the three Cagliari papers overlap in participants cannot be checked. For the 2024 row of Nolden and colleagues, we judged from the procedure and numbers that the 2020 paper on the first session belongs to the same series of experiments with overlapping participants, so the 2020 paper is not counted as an independent replication and is not given rows (participants were not matched individually). [rows: Nolden 2024 3.2 mM, PAV/PAV vs AVI/AVI; Melis 2013 3.2 mM, PAV/PAV vs AVI/AVI; Melis 2015 3.2 mM, PAV/PAV vs AVI/AVI; Sollai 2017 Paper disk, PAV/PAV vs AVI/AVI]
- Scales (gLMS, LMS), summaries (mean, median, geometric mean) and presentation differ between studies, so the sizes of values cannot be compared or turned into ratios across studies. For Campbell and colleagues, test results for pairs of genotypes cannot be confirmed in the text obtained and checked. Readings from figures carry some margin. [rows: Campbell 2012 PTC dilution score, PAV/PAV vs AVI/AVI; Bufe 2005 PTC 5600 µM, PAV/PAV vs AVI/AVI]
- Only studies rating standard PROP or PTC stimuli are included. Whether the bitterness of vegetables such as broccoli or kale, or of other foods, differs by genotype cannot be said from this table. Nor can differences in group averages decide each person's perception of bitterness or food preferences. [rows: Fischer 2014 Paper disk, PAV/PAV vs AVI/AVI; Nolden 2024 3.2 mM, PAV/PAV vs AVI/AVI]
How studies and rows were counted
The number of evidence rows is not the number of studies. Table rows, ledger evidence rows, papers and research groups are counted separately. Several rows come from one study, one per measure and genotype pair; the three pair rows of a study share participants and are not independent. The three papers from the same laboratory at the University of Cagliari are counted as one research group in the ledger; whether their participants overlap cannot be checked in the sources. Another paper on the first-session participants of the same study (the earlier paper by Nolden and colleagues) was judged from its procedure and numbers to belong to the same series of experiments with overlapping participants, so it is not counted as an independent replication and is not given rows. Three studies read in full but not tabulated (Garneau and colleagues and Behrens and colleagues, with no group values, and Mennella and colleagues, with only shares answering bitter at the same concentration) are not counted as tabulated sources and are described in the scope table. The "rows in the table" shown below are, on this page, the number of study × measured result × genotype pair.
- Sources in the table: 11, of which 11 have results on people's bitterness ratings in this table; for the others, only results on something else (compounds, smell) are shown, or they are used only for context.
- Rows in the table (study × measured result): 50. Rows on perceived bitterness: 50.
- Evidence rows in the lab's ledger these rows rest on: 0. One paper can supply several rows. Rows without a ledger evidence row come from other tables of the same papers, or from sources newly read for this question.
- Research groups among the studies with bitterness results in this table, as grouped in the ledger: 9 (reports from the same lab count as 1; co-authors and affiliations can overlap, so this is not a count of independent replications).
What we looked at
The sources looked at for this question and the reasons for including or excluding them are recorded here. We reviewed PROP, PTC and TAS2R38 sources in the lab's ledger, searched PubMed combining PROP, PTC, TAS2R38 and genotype terms, and saved and read candidates with full text from the official PMC pages. One paper whose publisher page could not be retrieved was read in the publisher-version PDF posted by the authors' laboratory. Studies readable only as abstracts were not tabulated and are on hold, because methods, numbers per group and comparison conditions cannot be checked in the full text. Studies with only groups defined by PROP ratings, studies of samples other than PROP or PTC (quinine, vegetables, foods and so on), studies without measured genotypes, and heritability studies are listed as out of scope with reasons. The search described here is a record and does not show that nothing was missed.
| Source | Decision | Why |
|---|---|---|
| The molecular basis of individual differences in phenylthiocarbamide and propylthiouracil bitterness perception (2005) doi | Included in the table | Recognition thresholds and suprathreshold intensity for PTC and PROP solutions compared across 3 genotype groups; cell assays are the main part of the paper, and the 3 individuals tested in detail (one each) are not given rows |
| Bitter Receptor Gene (TAS2R38), 6-n-Propylthiouracil (PROP) Bitterness and Alcohol Intake (2004) doi | Included in the table | PROP solution thresholds and intensity compared across 3 genotype groups; the same people were also grouped by PROP ratings, but pairs between those groups are not included, and only the correspondence with genotype (Table 2) is noted; alcohol results are not used |
| Age modifies the genotype-phenotype relationship for the bitter receptor TAS2R38 (2010) doi | Read as context (not in the table) | Read in full but not tabulated. PROP at 56, 180 and 560 µM was tasted in ascending order and people were classified by the first concentration called bitter (980 people). Among adults (PAV/PAV 75, PAV/AVI 164, AVI/AVI 78), the shares answering bitter at the weakest, 56 µM, were about 88%, 52% and 13% (approximate, Fig. 1). These are shares answering bitter at the same 56 µM, not a comparison of group concentration thresholds (most PAV/PAV and PAV/AVI people already answered at the weakest 56 µM). Tests are a genotype main effect and a genotype × age interaction in a multinomial logistic regression; tests between pairs of genotypes cannot be confirmed. Age differences appeared only within PAV/AVI. Most adults were mothers of child participants |
| The Associations Between 6-N-propylthiouracil (PROP) Intensity and Taste Intensities Differ by TAS2R38 Haplotype (2014) doi | Included in the table | Intensity of a PROP-impregnated filter-paper disk compared across 3 genotype groups (not a solution); correlations with the other 4 tastes are not used |
| Genetic, functional, and phenotypic diversity in TAS2R38-mediated bitter taste perception (2013) doi | Read as context (not in the table) | Read in full but not tabulated. For detection thresholds of PTC and PROP solutions (14 doubling steps; 48 people with the 3 main genotypes: PAV/PAV 12, PAV/AVI 27, AVI/AVI 9) there is no table of group values; the text says PAV/PAV thresholds were lowest, PAV/AVI intermediate and AVI/AVI highest (Fig. 7 shows distributions by step). The test is a regression using the number of PAV alleles carried (0, 1 or 2): PTC P=2.22×10⁻¹⁰ (r²=0.59), PROP P=6.44×10⁻⁹ (r²=0.52). This tests the association along the genotype order, not a pair of groups. Of the 56 participants, 28 also took part in a 2010 study from the same laboratory. The full text was read in the publisher-version PDF posted by the authors' laboratory. Thresholds for vegetable compounds such as goitrin and sinigrin are out of scope |
| Crowdsourcing taste research: genetic and phenotypic predictors of bitter taste perception as a model (2014) doi | Read as context (not in the table) | Read in full but not tabulated. For the intensity of a filter-paper disk impregnated with 0.453 M PROP (394 people), group rating values are not given; in the Table 1 multiple regression (log10(rating + 1), including sex, age and fungiform papillae) the coefficients relative to AVI/AVI were PAV/PAV +0.7818 and PAV/AVI +0.7221 (both p<0.0001). The coefficients are adjusted differences from the reference group, not group means. Separately, a t-test of the 2 PAV-carrying groups combined vs AVI/AVI gave p<0.0001. Fig. 2 was obtained, but its axis values do not match the description in the methods, so group ratings for the 3 groups cannot be fixed from it |
| The Gustin (CA6) Gene Polymorphism, rs2274333 (A/G), as a Mechanistic Link between PROP Tasting and Fungiform Taste Papilla Density and Maintenance (2013) doi | Included in the table | PROP solution thresholds and intensity compared across 3 TAS2R38 groups; gustin genotype comparisons are not used; one of the three Cagliari papers |
| Dose-Dependent Effects of L-Arginine on PROP Bitterness Intensity and Latency and Characteristics of the Chemical Interaction between PROP and L-Arginine (2015) doi | Included in the table | Intensity of plain PROP solution compared across 3 genotype groups; solutions with added L-arginine and groups defined by PROP ratings are not used; one of the three Cagliari papers |
| First objective evaluation of taste sensitivity to 6-n-propylthiouracil (PROP), a paradigm gustatory stimulus in humans (2017) doi | Included in the table | Intensity of a PROP-impregnated paper disk compared across 3 genotype groups; electrophysiology, fungiform papilla results and groups defined by PROP ratings are not used; one of the three Cagliari papers |
| Differential Activation of TAS2R4 May Recover Ability to Taste Propylthiouracil for Some TAS2R38 AVI Homozygotes (2024) doi | Included in the table | PROP solution intensity compared across 3 genotype groups (approximate figure readings); TAS2R4 and cell assays are not used |
| Association between Genetic Variation in the TAS2R38 Bitter Taste Receptor and Propylthiouracil Bitter Taste Thresholds among Adults Living in Japan Using the Modified 2AFC Procedure with the Quest Method (2023) doi | Included in the table | PROP solution detection thresholds compared across 3 genotype groups (adults in Japan) |
| Evolution of Functionally Diverse Alleles Associated with PTC Bitter Taste Sensitivity in Africa (2012) doi | Included in the table | PTC dilution scores compared by genotype (African populations); pairwise tests cannot be confirmed |
| The Associations Between Bitter and Fat Taste Sensitivity, and Dietary Fat Intake: Are They Impacted by Genetic Predisposition? (2021) doi | Included in the table | PTC strip intensity compared by genotype; fat taste and dietary results are not used |
| Validation of Edible Taste Strips for Assessing PROP Taste Perception (2013) doi | Included in the table | Intensity of PROP solutions and dissolvable films compared across 3 genotype groups in the same participants; liking and taste-quality results are not used |
| Predominant Qualities Evoked by Quinine, Sucrose, and Capsaicin Associate With PROP Bitterness, but not TAS2R38 Genotype (2020) doi | Read as context (not in the table) | Judged from its procedure and numbers (identical PAV/PAV and AVI/AVI counts) to cover first-session participants of the same series of experiments as Nolden 2024 in the table; not counted as an independent replication, so not given rows and only mentioned in the note to the 2024 row (2020 values are not moved into the 2024 row) |
| Worldwide study of the taste of bitter medicines and their modifiers (2025) doi | Read as context (not in the table) | On hold. A 560 µM PROP solution was rated as a control, and the text gives the order of diplotype means (AVI/AVI < AVI/PAV < PAV/PAV), but neither group values nor test methods. Fig. 6 shows 3 groups of a single-SNP genotype (rs713598), not diplotypes, so its tests are not moved to diplotype pairs |
| Supertasting and PROP bitterness depends on more than the TAS2R38 gene (2008) doi | Read as context (not in the table) | On hold. Only the abstract could be read, so the methods, numbers per group, comparison conditions and pairwise tests cannot be checked in the full text; not tabulated. The publisher page returned a verification screen |
| TAS2R38 genotypes and phenylthiocarbamide bitter taste perception in a population of young adults (2009) doi | Read as context (not in the table) | On hold. Only the abstract could be read, so the methods, numbers per group, comparison conditions and pairwise tests cannot be checked in the full text; not tabulated. The abstract gives mean 9-point ratings for PTC strips by a single site (position 49: AA, AP, PP), not by 3-site diplotypes |
| Polymorphisms in TAS2R38 and the taste bud trophic factor, gustin gene co-operate in modulating PROP taste phenotype (2011) doi | Read as context (not in the table) | On hold. Only the abstract could be read, so the methods, numbers per group, comparison conditions and pairwise tests cannot be checked in the full text; not tabulated. A Cagliari study whose participants may overlap with the three in the table |
| Positional cloning of the human quantitative trait locus underlying taste sensitivity to phenylthiocarbamide (2003) doi | Read as context (not in the table) | On hold. Only the abstract could be read, so the methods, numbers per group, comparison conditions and pairwise tests cannot be checked in the full text; not tabulated. The publisher full text is by subscription and the abstract gives no group values |
| Variations in the TAS2R38 gene among college students in Hubei (2022) doi | Read as context (not in the table) | On hold. The correspondence between PTC threshold classes and diplotypes is in a supplementary table that was not obtained; the text only says "strong correlation (P<0.01)", without group values or pairwise tests |
| Vegetable Intake in College-Aged Adults Is Explained by Oral Sensory Phenotypes and TAS2R38 Genotype (2010) doi | Read as context (not in the table) | On hold (an additional candidate whose source was not checked). It is listed in the scope of the page on brassica bitterness and TAS2R38 as a study that measured PROP solution bitterness; the source was not read for this question |
| Binary taste mixture interactions in PROP non-tasters, medium-tasters and super-tasters (2001) doi | Excluded | Out of scope: the samples rated were mixtures of sucrose and NaCl with quinine and citric acid, not PROP or PTC pairs (groups were also defined by PROP ratings) |
| PTC/PROP tasting: anatomy, psychophysics, and sex effects (1994) doi | Excluded | Out of scope: deals with rating-defined groups and fungiform papillae, with no genotype-group pairs (abstract only) |
| Chemesthesis and taste: Evidence of independent processing of sensation intensity (2005) doi | Excluded | Out of scope: groups defined by whether temperature alone evokes taste, not PROP/PTC genotype groups (abstract only) |
| Specific alleles of bitter receptor genes influence human sensitivity to the bitterness of aloin and saccharin (2007) doi | Excluded | Out of scope: other receptor genes and the bitterness of aloin and saccharin, not PROP or PTC |
| Heritability and Genetic Covariation of Sensitivity to PROP, SOA, Quinine HCl, and Caffeine (2006) doi | Excluded | Out of scope: a twin study estimating heritability of PROP bitterness, with no TAS2R38 genotype-group pairs |
| Taste Perception of Sweet, Sour, Salty, Bitter, and Umami and Changes Due to l-Arginine Supplementation, as a Function of Genetic Ability to Taste 6-n-Propylthiouracil (2017) doi | Excluded | Out of scope: PROP was used to classify groups by rating, and the comparisons concern other basic tastes and changes with added L-arginine |
| Prevalence and Genetic Analysis of Bitter Taste Perception for Phenylthiocarbamide (PTC) Among Some Muslim Populations of Uttar Pradesh, India (2014) PubMed | Excluded | Out of scope: counts the share of PTC tasters in populations in Uttar Pradesh, India, without comparing groups by measured TAS2R38 genotype (published in an Iranian public-health journal) |
| Consumption of lingonberries by TAS2R38 genotype and sensory quality of texture-designed lingonberry samples (2015) doi | Excluded | Out of scope: a food (lingonberry) is the sample, not a standard PROP or PTC stimulus |
| The Effects of Cranberry Polyphenol Extract (CPE) Supplementation on Astringency and Flavor Perception as a Function of PROP Taster Status and Other Individual Factors (2022) doi | Excluded | Out of scope: a food extract (cranberry polyphenols) is the sample, and groups were defined by PROP ratings |
Source ledger
The source of each study in the tables can be traced from its section. The lab's full list of sources is in the source ledger (in Japanese).
How to cite this page
Aoki, M. (NIGAMI LAB) (2026). Supertasters, PROP/PTC and TAS2R38 genotype. https://nigamilab.com/en/topics/prop-ptc-tas2r38-bitterness/
Each claim card has a stable anchor (#clm-…). The dataset is CC BY 4.0; see the data page. Dataset DOI: 10.5281/zenodo.22739101