Predicting human olfactory perception from chemical features of odor molecules

Andreas Keller; Richard C. Gerkin; Yuanfang Guan; Amit Dhurandhar; Gabor Turu; Bence Szalai; Joel D. Mainland; Yusuke Ihara; Chung Wen Yu; Russ Wolfinger; Celine Vens; Leander Schietgat; Kurt De Grave; Raquel Norel; Gustavo Stolovitzky; Guillermo A. Cecchi; Leslie B. Vosshall; Pablo Meyer; A. P. Bhondekar; P. C. Boutros; Y. C. Chang; C. Y. Chen; B. W. Cherng; A. Dimitriev; A. Dolenc; A. O. Falcao; A. K. Golińska; M. Y. Hong; P. H. Hsieh; B. F. Huang; L. Hunyady; R. Kaur; M. D. Kazanov; R. Kumar; W. Lesiński; X. Lin; A. Matteson; Y. J. Oyang; B. Panwar; R. Piliszek; A. Polewko-Klim; G. P.S. Raghava; W. R. Rudnicki; L. Saiz; R. X. Sun; M. Toplak; Y. A. Tung; P. Us; P. Várnai; JMG Vilar; M. Xie; D. Yao; M. Zitnik; B. Zupan

doi:10.1126/science.aal2014

Predicting human olfactory perception from chemical features of odor molecules

Andreas Keller
, Richard C. Gerkin
, Yuanfang Guan
, Amit Dhurandhar
, Gabor Turu
, Bence Szalai
, Joel D. Mainland
, Yusuke Ihara
, Chung Wen Yu
, Russ Wolfinger
, Celine Vens
, Leander Schietgat
, Kurt De Grave
, Raquel Norel
, Gustavo Stolovitzky
, Guillermo A. Cecchi
, Leslie B. Vosshall
, Pablo Meyer
, A. P. Bhondekar
, P. C. Boutros

Y. C. Chang, C. Y. Chen, B. W. Cherng, A. Dimitriev, A. Dolenc, A. O. Falcao, A. K. Golińska, M. Y. Hong, P. H. Hsieh, B. F. Huang, L. Hunyady, R. Kaur, M. D. Kazanov, R. Kumar, W. Lesiński, X. Lin, A. Matteson, Y. J. Oyang, B. Panwar, R. Piliszek, A. Polewko-Klim, G. P.S. Raghava, W. R. Rudnicki, L. Saiz, R. X. Sun, M. Toplak, Y. A. Tung, P. Us, P. Várnai, JMG Vilar, M. Xie, D. Yao, M. Zitnik, B. Zupan

Research output: Contribution to journal › Article › peer-review

250 Scopus citations

Abstract

It is still not possible to predict whether a given molecule will have a perceived odor or what olfactory percept it will produce.We therefore organized the crowd-sourced DREAM Olfaction Prediction Challenge. Using a large olfactory psychophysical data set, teams developed machine-learning algorithms to predict sensory attributes of molecules based on their chemoinformatic features.The resulting models accurately predicted odor intensity and pleasantness and also successfully predicted 8 among 19 rated semantic descriptors ("garlic," "fish," "sweet," "fruit," "burnt," "spices," "flower," and "sour"). Regularized linear models performed nearly as well as random forest-based ones, with a predictive accuracy that closely approaches a key theoretical limit.These models help to predict the perceptual qualities of virtually any molecule with high accuracy and also reverse-engineer the smell of a molecule.

Original language	English
Pages (from-to)	820-826
Number of pages	7
Journal	Science
Volume	355
Issue number	6327
DOIs	https://doi.org/10.1126/science.aal2014
State	Published - 24 Feb 2017
Externally published	Yes

Access to Document

10.1126/science.aal2014

Cite this

Keller, A., Gerkin, R. C., Guan, Y., Dhurandhar, A., Turu, G., Szalai, B., Mainland, J. D., Ihara, Y., Yu, C. W., Wolfinger, R., Vens, C., Schietgat, L., De Grave, K., Norel, R., Stolovitzky, G., Cecchi, G. A., Vosshall, L. B., Meyer, P., Bhondekar, A. P., ... Zupan, B. (2017). Predicting human olfactory perception from chemical features of odor molecules. Science, 355(6327), 820-826. https://doi.org/10.1126/science.aal2014

@article{0e0586b575cc4c878bc2ab3f28a429ee,

title = "Predicting human olfactory perception from chemical features of odor molecules",

abstract = "It is still not possible to predict whether a given molecule will have a perceived odor or what olfactory percept it will produce.We therefore organized the crowd-sourced DREAM Olfaction Prediction Challenge. Using a large olfactory psychophysical data set, teams developed machine-learning algorithms to predict sensory attributes of molecules based on their chemoinformatic features.The resulting models accurately predicted odor intensity and pleasantness and also successfully predicted 8 among 19 rated semantic descriptors ({"}garlic,{"} {"}fish,{"} {"}sweet,{"} {"}fruit,{"} {"}burnt,{"} {"}spices,{"} {"}flower,{"} and {"}sour{"}). Regularized linear models performed nearly as well as random forest-based ones, with a predictive accuracy that closely approaches a key theoretical limit.These models help to predict the perceptual qualities of virtually any molecule with high accuracy and also reverse-engineer the smell of a molecule.",

author = "Andreas Keller and Gerkin, \{Richard C.\} and Yuanfang Guan and Amit Dhurandhar and Gabor Turu and Bence Szalai and Mainland, \{Joel D.\} and Yusuke Ihara and Yu, \{Chung Wen\} and Russ Wolfinger and Celine Vens and Leander Schietgat and \{De Grave\}, Kurt and Raquel Norel and Gustavo Stolovitzky and Cecchi, \{Guillermo A.\} and Vosshall, \{Leslie B.\} and Pablo Meyer and Bhondekar, \{A. P.\} and Boutros, \{P. C.\} and Chang, \{Y. C.\} and Chen, \{C. Y.\} and Cherng, \{B. W.\} and A. Dimitriev and A. Dolenc and Falcao, \{A. O.\} and Goli{\'n}ska, \{A. K.\} and Hong, \{M. Y.\} and Hsieh, \{P. H.\} and Huang, \{B. F.\} and L. Hunyady and R. Kaur and Kazanov, \{M. D.\} and R. Kumar and W. Lesi{\'n}ski and X. Lin and A. Matteson and Oyang, \{Y. J.\} and B. Panwar and R. Piliszek and A. Polewko-Klim and Raghava, \{G. P.S.\} and Rudnicki, \{W. R.\} and L. Saiz and Sun, \{R. X.\} and M. Toplak and Tung, \{Y. A.\} and P. Us and P. V{\'a}rnai and JMG Vilar and M. Xie and D. Yao and M. Zitnik and B. Zupan",

year = "2017",

month = feb,

day = "24",

doi = "10.1126/science.aal2014",

language = "English",

volume = "355",

pages = "820--826",

journal = "Science",

issn = "0036-8075",

publisher = "American Association for the Advancement of Science",

number = "6327",

}

Keller, A, Gerkin, RC, Guan, Y, Dhurandhar, A, Turu, G, Szalai, B, Mainland, JD, Ihara, Y, Yu, CW, Wolfinger, R, Vens, C, Schietgat, L, De Grave, K, Norel, R, Stolovitzky, G, Cecchi, GA, Vosshall, LB, Meyer, P, Bhondekar, AP, Boutros, PC, Chang, YC, Chen, CY, Cherng, BW, Dimitriev, A, Dolenc, A, Falcao, AO, Golińska, AK, Hong, MY, Hsieh, PH, Huang, BF, Hunyady, L, Kaur, R, Kazanov, MD, Kumar, R, Lesiński, W, Lin, X, Matteson, A, Oyang, YJ, Panwar, B, Piliszek, R, Polewko-Klim, A, Raghava, GPS, Rudnicki, WR, Saiz, L, Sun, RX, Toplak, M, Tung, YA, Us, P, Várnai, P, Vilar, JMG, Xie, M, Yao, D, Zitnik, M & Zupan, B 2017, 'Predicting human olfactory perception from chemical features of odor molecules', Science, vol. 355, no. 6327, pp. 820-826. https://doi.org/10.1126/science.aal2014

TY - JOUR

T1 - Predicting human olfactory perception from chemical features of odor molecules

AU - Keller, Andreas

AU - Gerkin, Richard C.

AU - Guan, Yuanfang

AU - Dhurandhar, Amit

AU - Turu, Gabor

AU - Szalai, Bence

AU - Mainland, Joel D.

AU - Ihara, Yusuke

AU - Yu, Chung Wen

AU - Wolfinger, Russ

AU - Vens, Celine

AU - Schietgat, Leander

AU - De Grave, Kurt

AU - Norel, Raquel

AU - Stolovitzky, Gustavo

AU - Cecchi, Guillermo A.

AU - Vosshall, Leslie B.

AU - Meyer, Pablo

AU - Bhondekar, A. P.

AU - Boutros, P. C.

AU - Chang, Y. C.

AU - Chen, C. Y.

AU - Cherng, B. W.

AU - Dimitriev, A.

AU - Dolenc, A.

AU - Falcao, A. O.

AU - Golińska, A. K.

AU - Hong, M. Y.

AU - Hsieh, P. H.

AU - Huang, B. F.

AU - Hunyady, L.

AU - Kaur, R.

AU - Kazanov, M. D.

AU - Kumar, R.

AU - Lesiński, W.

AU - Lin, X.

AU - Matteson, A.

AU - Oyang, Y. J.

AU - Panwar, B.

AU - Piliszek, R.

AU - Polewko-Klim, A.

AU - Raghava, G. P.S.

AU - Rudnicki, W. R.

AU - Saiz, L.

AU - Sun, R. X.

AU - Toplak, M.

AU - Tung, Y. A.

AU - Us, P.

AU - Várnai, P.

AU - Vilar, JMG

AU - Xie, M.

AU - Yao, D.

AU - Zitnik, M.

AU - Zupan, B.

PY - 2017/2/24

Y1 - 2017/2/24

N2 - It is still not possible to predict whether a given molecule will have a perceived odor or what olfactory percept it will produce.We therefore organized the crowd-sourced DREAM Olfaction Prediction Challenge. Using a large olfactory psychophysical data set, teams developed machine-learning algorithms to predict sensory attributes of molecules based on their chemoinformatic features.The resulting models accurately predicted odor intensity and pleasantness and also successfully predicted 8 among 19 rated semantic descriptors ("garlic," "fish," "sweet," "fruit," "burnt," "spices," "flower," and "sour"). Regularized linear models performed nearly as well as random forest-based ones, with a predictive accuracy that closely approaches a key theoretical limit.These models help to predict the perceptual qualities of virtually any molecule with high accuracy and also reverse-engineer the smell of a molecule.

AB - It is still not possible to predict whether a given molecule will have a perceived odor or what olfactory percept it will produce.We therefore organized the crowd-sourced DREAM Olfaction Prediction Challenge. Using a large olfactory psychophysical data set, teams developed machine-learning algorithms to predict sensory attributes of molecules based on their chemoinformatic features.The resulting models accurately predicted odor intensity and pleasantness and also successfully predicted 8 among 19 rated semantic descriptors ("garlic," "fish," "sweet," "fruit," "burnt," "spices," "flower," and "sour"). Regularized linear models performed nearly as well as random forest-based ones, with a predictive accuracy that closely approaches a key theoretical limit.These models help to predict the perceptual qualities of virtually any molecule with high accuracy and also reverse-engineer the smell of a molecule.

UR - https://www.scopus.com/pages/publications/85013999403

U2 - 10.1126/science.aal2014

DO - 10.1126/science.aal2014

M3 - Article

C2 - 28219971

AN - SCOPUS:85013999403

SN - 0036-8075

VL - 355

SP - 820

EP - 826

JO - Science

JF - Science

IS - 6327

ER -

Predicting human olfactory perception from chemical features of odor molecules

Abstract

Access to Document

Fingerprint

Cite this