Machine learning overcomes human bias in the discovery of self-assembling peptides

Rohit Batra; Troy D. Loeffler; Henry Chan; Srilok Srinivasan; Honggang Cui; Ivan V. Korendovych; Vikas Nanda; Liam C. Palmer; Lee A. Solomon; H. Christopher Fry; Subramanian K.R.S. Sankaranarayanan

doi:10.1038/s41557-022-01055-3

Machine learning overcomes human bias in the discovery of self-assembling peptides

Rohit Batra, Troy D. Loeffler, Henry Chan, Srilok Srinivasan, Honggang Cui, Ivan V. Korendovych, Vikas Nanda, Liam C. Palmer, Lee A. Solomon, H. Christopher Fry^*, Subramanian K.R.S. Sankaranarayanan^*

^*Corresponding author for this work

CRN - Center for Regenerative Nanomedicine

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

63 Scopus citations

Abstract

Peptide materials have a wide array of functions, from tissue engineering and surface coatings to catalysis and sensing. Tuning the sequence of amino acids that comprise the peptide modulates peptide functionality, but a small increase in sequence length leads to a dramatic increase in the number of peptide candidates. Traditionally, peptide design is guided by human expertise and intuition and typically yields fewer than ten peptides per study, but these approaches are not easily scalable and are susceptible to human bias. Here we introduce a machine learning workflow—AI-expert—that combines Monte Carlo tree search and random forest with molecular dynamics simulations to develop a fully autonomous computational search engine to discover peptide sequences with high potential for self-assembly. We demonstrate the efficacy of the AI-expert to efficiently search large spaces of tripeptides and pentapeptides. The predictability of AI-expert performs on par or better than our human experts and suggests several non-intuitive sequences with high self-assembly propensity, outlining its potential to overcome human bias and accelerate peptide discovery. [Figure not available: see fulltext.]

Original language	English (US)
Pages (from-to)	1427-1435
Number of pages	9
Journal	Nature chemistry
Volume	14
Issue number	12
DOIs	https://doi.org/10.1038/s41557-022-01055-3
State	Published - Dec 2022

Funding

Work performed at the Center for Nanoscale Materials, a US Department of Energy (DOE) Office of Science User Facility, was supported by the US DOE, Office of Basic Energy Sciences, under contract no. DE-AC02-06CH11357, and additionally supported by the University of Chicago and the DOE under DOE contract no. DE-AC02-06CH11357 awarded to UChicago Argonne, LLC, operator of the Argonne National Laboratory. This material is based on work supported by the DOE, Office of Science, BES Data, Artificial Intelligence and Machine Learning at DOE Scientific User Facilities programme (Digital Twins). We gratefully acknowledge the computing resources provided on Bebop, the high-performance computing clusters operated by the Laboratory Computing Resource Center (LCRC) at Argonne National Laboratory. S.K.R.S.S. acknowledges support from the UIC faculty start-up fund. We acknowledge T. Tuttle for sharing computational data on tripeptides.

ASJC Scopus subject areas

General Chemistry
General Chemical Engineering

Access to Document

10.1038/s41557-022-01055-3

Cite this

@article{7e1c1702fd8f464f93b1abac0e87a409,

title = "Machine learning overcomes human bias in the discovery of self-assembling peptides",

abstract = "Peptide materials have a wide array of functions, from tissue engineering and surface coatings to catalysis and sensing. Tuning the sequence of amino acids that comprise the peptide modulates peptide functionality, but a small increase in sequence length leads to a dramatic increase in the number of peptide candidates. Traditionally, peptide design is guided by human expertise and intuition and typically yields fewer than ten peptides per study, but these approaches are not easily scalable and are susceptible to human bias. Here we introduce a machine learning workflow—AI-expert—that combines Monte Carlo tree search and random forest with molecular dynamics simulations to develop a fully autonomous computational search engine to discover peptide sequences with high potential for self-assembly. We demonstrate the efficacy of the AI-expert to efficiently search large spaces of tripeptides and pentapeptides. The predictability of AI-expert performs on par or better than our human experts and suggests several non-intuitive sequences with high self-assembly propensity, outlining its potential to overcome human bias and accelerate peptide discovery. [Figure not available: see fulltext.]",

author = "Rohit Batra and Loeffler, {Troy D.} and Henry Chan and Srilok Srinivasan and Honggang Cui and Korendovych, {Ivan V.} and Vikas Nanda and Palmer, {Liam C.} and Solomon, {Lee A.} and Fry, {H. Christopher} and Sankaranarayanan, {Subramanian K.R.S.}",

note = "Publisher Copyright: {\textcopyright} 2022, The Author(s), under exclusive licence to Springer Nature Limited.",

year = "2022",

month = dec,

doi = "10.1038/s41557-022-01055-3",

language = "English (US)",

volume = "14",

pages = "1427--1435",

journal = "Nature chemistry",

issn = "1755-4330",

publisher = "Nature Publishing Group",

number = "12",

}

TY - JOUR

T1 - Machine learning overcomes human bias in the discovery of self-assembling peptides

AU - Batra, Rohit

AU - Loeffler, Troy D.

AU - Chan, Henry

AU - Srinivasan, Srilok

AU - Cui, Honggang

AU - Korendovych, Ivan V.

AU - Nanda, Vikas

AU - Palmer, Liam C.

AU - Solomon, Lee A.

AU - Fry, H. Christopher

AU - Sankaranarayanan, Subramanian K.R.S.

PY - 2022/12

Y1 - 2022/12

N2 - Peptide materials have a wide array of functions, from tissue engineering and surface coatings to catalysis and sensing. Tuning the sequence of amino acids that comprise the peptide modulates peptide functionality, but a small increase in sequence length leads to a dramatic increase in the number of peptide candidates. Traditionally, peptide design is guided by human expertise and intuition and typically yields fewer than ten peptides per study, but these approaches are not easily scalable and are susceptible to human bias. Here we introduce a machine learning workflow—AI-expert—that combines Monte Carlo tree search and random forest with molecular dynamics simulations to develop a fully autonomous computational search engine to discover peptide sequences with high potential for self-assembly. We demonstrate the efficacy of the AI-expert to efficiently search large spaces of tripeptides and pentapeptides. The predictability of AI-expert performs on par or better than our human experts and suggests several non-intuitive sequences with high self-assembly propensity, outlining its potential to overcome human bias and accelerate peptide discovery. [Figure not available: see fulltext.]

AB - Peptide materials have a wide array of functions, from tissue engineering and surface coatings to catalysis and sensing. Tuning the sequence of amino acids that comprise the peptide modulates peptide functionality, but a small increase in sequence length leads to a dramatic increase in the number of peptide candidates. Traditionally, peptide design is guided by human expertise and intuition and typically yields fewer than ten peptides per study, but these approaches are not easily scalable and are susceptible to human bias. Here we introduce a machine learning workflow—AI-expert—that combines Monte Carlo tree search and random forest with molecular dynamics simulations to develop a fully autonomous computational search engine to discover peptide sequences with high potential for self-assembly. We demonstrate the efficacy of the AI-expert to efficiently search large spaces of tripeptides and pentapeptides. The predictability of AI-expert performs on par or better than our human experts and suggests several non-intuitive sequences with high self-assembly propensity, outlining its potential to overcome human bias and accelerate peptide discovery. [Figure not available: see fulltext.]

UR - http://www.scopus.com/inward/record.url?scp=85140960548&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85140960548&partnerID=8YFLogxK

U2 - 10.1038/s41557-022-01055-3

DO - 10.1038/s41557-022-01055-3

M3 - Article

C2 - 36316409

AN - SCOPUS:85140960548

SN - 1755-4330

VL - 14

SP - 1427

EP - 1435

JO - Nature chemistry

JF - Nature chemistry

IS - 12

ER -

Machine learning overcomes human bias in the discovery of self-assembling peptides

Abstract

Funding

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this