Molecular representations in AI-driven drug discovery: a review and practical guide

Journal of Cheminformatics, Journal Year: 2020, Volume and Issue: 12(1)

Published: Sept. 17, 2020

The technological advances of the past century, marked by computer revolution and advent high-throughput screening technologies in drug discovery, opened path to computational analysis visualization bioactive molecules. For this purpose, it became necessary represent molecules a syntax that would be readable computers understandable scientists various fields. A large number chemical representations have been developed over years, their numerosity being due fast development complexity producing representation encompasses all structural characteristics. We present here some most popular electronic molecular macromolecular used many which are based on graph representations. Furthermore, we describe applications these AI-driven discovery. Our aim is provide brief guide essential practice AI This review serves as for researchers who little experience with handling plan work at interface

Language: Английский

---

Machine learning the quantum-chemical properties of metal–organic frameworks for accelerated materials discovery

Matter, Journal Year: 2021, Volume and Issue: 4(5), P. 1578 - 1597

Published: April 5, 2021

The modular nature of metal–organic frameworks (MOFs) enables synthetic control over their physical and chemical properties, but it can be difficult to know which MOFs would optimal for a given application. High-throughput computational screening machine learning are promising routes efficiently navigate the vast space have rarely been used prediction properties that need calculated by quantum mechanical methods. Here, we introduce Quantum MOF (QMOF) database, publicly available database computed quantum-chemical more than 14,000 experimentally synthesized MOFs. Throughout this study, demonstrate how models trained on QMOF rapidly discover with targeted electronic structure using theoretically band gaps as representative example. We conclude highlighting several predicted low gaps, challenging task electronically insulating most

Language: Английский

---

Bimetallic Sites for Catalysis: From Binuclear Metal Sites to Bimetallic Nanoclusters and Nanoparticles

Chemical Reviews, Journal Year: 2023, Volume and Issue: 123(8), P. 4855 - 4933

Published: March 27, 2023

Heterogeneous bimetallic catalysts have broad applications in industrial processes, but achieving a fundamental understanding on the nature of active sites at atomic and molecular level is very challenging due to structural complexity catalysts. Comparing features catalytic performances different entities will favor formation unified structure-reactivity relationships heterogeneous thereby facilitate upgrading current In this review, we discuss geometric electronic structures three representative types (bimetallic binuclear sites, nanoclusters, nanoparticles) then summarize synthesis methodologies characterization techniques for entities, with emphasis recent progress made past decade. The supported nanoparticles series important reactions are discussed. Finally, future research directions catalysis based and, more generally, prospective developments both practical applications.

Language: Английский

---

Computational Discovery of Transition-metal Complexes: From High-throughput Screening to Machine Learning

Chemical Reviews, Journal Year: 2021, Volume and Issue: 121(16), P. 9927 - 10000

Published: July 14, 2021

Transition-metal complexes are attractive targets for the design of catalysts and functional materials. The behavior metal-organic bond, while very tunable achieving target properties, is challenging to predict necessitates searching a wide complex space identify needles in haystacks applications. This review will focus on techniques that make high-throughput search transition-metal chemical feasible discovery with desirable properties. cover development, promise, limitations "traditional" computational chemistry (i.e., force field, semiempirical, density theory methods) as it pertains data generation inorganic molecular discovery. also discuss opportunities leveraging experimental sources. We how advances statistical modeling, artificial intelligence, multiobjective optimization, automation accelerate lead compounds rules. overall objective this showcase bringing together from diverse areas computer science have enabled rapid uncovering structure-property relationships chemistry. aim highlight unique considerations motifs bonding (e.g., variable spin oxidation state, strength/nature) set them their apart more commonly considered organic molecules. uncertainty relative scarcity motivate specific developments machine learning representations, model training, Finally, we conclude an outlook opportunity accelerated complexes.

Language: Английский

---

A field guide to flow chemistry for synthetic organic chemists

Chemical Science, Journal Year: 2023, Volume and Issue: 14(16), P. 4230 - 4247

Published: Jan. 1, 2023

This review explores the benefits of flow chemistry and dispels notion that it is a mysterious “black box”, demonstrating how can push boundaries organic synthesis through understanding its governing principles.

Language: Английский

---

Deep Learning in Protein Structural Modeling and Design

Patterns, Journal Year: 2020, Volume and Issue: 1(9), P. 100142 - 100142

Published: Nov. 12, 2020

Deep learning is catalyzing a scientific revolution fueled by big data, accessible toolkits, and powerful computational resources, impacting many fields, including protein structural modeling. Protein modeling, such as predicting structure from amino acid sequence evolutionary information, designing proteins toward desirable functionality, or properties behavior of protein, critical to understand engineer biological systems at the molecular level. In this review, we summarize recent advances in applying deep techniques tackle problems modeling design. We dissect emerging approaches using for discuss challenges that must be addressed. argue central importance structure, following "sequence → function" paradigm. This review directed help both biologists gain familiarity with methods applied computer scientists perspective on biologically meaningful may benefit techniques.

Language: Английский

---

Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems

Chemical Reviews, Journal Year: 2021, Volume and Issue: 121(16), P. 9816 - 9872

Published: July 7, 2021

Machine learning models are poised to make a transformative impact on chemical sciences by dramatically accelerating computational algorithms and amplifying insights available from chemistry methods. However, achieving this requires confluence coaction of expertise in computer science physical sciences. This Review is written for new experienced researchers working at the intersection both fields. We first provide concise tutorials machine methods, showing how involving can be achieved. follow with critical review noteworthy applications that demonstrate used together insightful (and useful) predictions molecular materials modeling, retrosyntheses, catalysis, drug design.

Language: Английский

---

Data-science driven autonomous process optimization

Communications Chemistry, Journal Year: 2021, Volume and Issue: 4(1)

Published: Aug. 2, 2021

Autonomous process optimization involves the human intervention-free exploration of a range parameters to improve responses such as product yield and selectivity. Utilizing off-the-shelf components, we develop closed-loop system for carrying out parallel autonomous experiments in batch. Upon implementation our stereoselective Suzuki-Miyaura coupling, find that definition set meaningful, broad, unbiased is most critical aspect successful optimization. Importantly, discern phosphine ligand, categorical parameter, vital determination reaction outcome. To date, parameter selection has relied on chemical intuition, potentially introducing bias into experimental design. In seeking systematic method selecting diverse ligands, strategy leverages computed molecular feature clustering. The resulting uncovers conditions selectively access desired isomer high yield.

Language: Английский

---

Generative Models as an Emerging Paradigm in the Chemical Sciences

Journal of the American Chemical Society, Journal Year: 2023, Volume and Issue: 145(16), P. 8736 - 8750

Published: April 13, 2023

Traditional computational approaches to design chemical species are limited by the need compute properties for a vast number of candidates, e.g., discriminative modeling. Therefore, inverse methods aim start from desired property and optimize corresponding structure. From machine learning viewpoint, problem can be addressed through so-called generative Mathematically, models defined probability distribution function given molecular or material In contrast, model seeks exploit joint with target characteristics. The overarching idea modeling is implement system that produces novel compounds expected have set features, effectively sidestepping issues found in forward process. this contribution, we overview critically analyze popular algorithms like adversarial networks, variational autoencoders, flow, diffusion models. We highlight key differences between each models, provide insights into recent success stories, discuss outstanding challenges realizing discovered solutions applications.

Language: Английский

---

Predicting Reaction Yields via Supervised Learning

Accounts of Chemical Research, Journal Year: 2021, Volume and Issue: 54(8), P. 1856 - 1865

Published: March 31, 2021

ConspectusNumerous disciplines, such as image recognition and language translation, have been revolutionized by using machine learning (ML) to leverage big data. In organic synthesis, providing accurate chemical reactivity predictions with supervised ML could assist chemists reaction prediction, optimization, mechanistic interrogation.To apply reactions, one needs define the object of prediction (e.g., yield, enantioselectivity, solubility, or a recommendation) represent reactions descriptive Our group's effort has focused on representing DFT-derived physical features reacting molecules conditions, which serve for building models.In this Account, we present review perspective three studies conducted our group where models employed predict yield. First, focus small data set 16 phosphine ligands were evaluated in single Ni-catalyzed Suzuki–Miyaura cross-coupling reaction, yield was modeled linear regression. setting, regression complexity is strongly limited amount available data, emphasize importance identifying that are directly relevant reactivity. Next, trained two larger sets obtained high-throughput experimentation (HTE). With hundreds thousands available, more complex can be explored, example, algorithmically perform feature selection from broad candidate features. We examine how variety algorithms model these well generalize out-of-sample substrates. Specifically, compare use DFT-based featurization baseline carry no information, is, random features, naive non-ML averages yields share same conditions substrate combinations. find only sets, leads significant, although moderate, improvement. The source improvement further isolated specific allowed us formulate testable hypothesis validated experimentally. Finally, offer remarks HTE focusing algorithmic improvements training.Statistical methods chemistry rich history, but recently gained widespread attention development. As untapped potential novel tools likely arise future research. suggest lead improved over simpler modeling facilitate understanding dynamics. However, research development required establish an indispensable tool modeling.

Language: Английский

---