Transfer Learning for Drug Discovery

Journal of Medicinal Chemistry, Год журнала: 2020, Номер 63(16), С. 8683 - 8694

Опубликована: Июль 16, 2020

The data sets available to train models for in silico drug discovery efforts are often small. Indeed, the sparse availability of labeled is a major barrier artificial-intelligence-assisted discovery. One solution this problem develop algorithms that can cope with relatively heterogeneous and scarce data. Transfer learning type machine leverage existing, generalizable knowledge from other related tasks enable separate task small set Deep transfer most commonly used field This Perspective provides an overview applications date. Furthermore, it outlooks on future development

Язык: Английский

---

The Hitchhiker's guide to biocatalysis: recent advances in the use of enzymes in organic synthesis

Chemical Science, Год журнала: 2020, Номер 11(10), С. 2587 - 2605

Опубликована: Янв. 1, 2020

Enzymes are excellent catalysts that increasingly being used in industry and academia. This Perspective provides a general practical guide to enzymes their synthetic potential, primarily aimed at organic chemists.

Язык: Английский

---

The rise of self-driving labs in chemical and materials sciences

Nature Synthesis, Год журнала: 2023, Номер 2(6), С. 483 - 492

Опубликована: Янв. 30, 2023

Язык: Английский

---

Artificial Chemist: An Autonomous Quantum Dot Synthesis Bot

Advanced Materials, Год журнала: 2020, Номер 32(30)

Опубликована: Июнь 4, 2020

Abstract The optimal synthesis of advanced nanomaterials with numerous reaction parameters, stages, and routes, poses one the most complex challenges modern colloidal science, current strategies often fail to meet demands these combinatorially large systems. In response, an Artificial Chemist is presented: integration machine‐learning‐based experiment selection high‐efficiency autonomous flow chemistry. With self‐driving Chemist, made‐to‐measure inorganic perovskite quantum dots (QDs) in are autonomously synthesized, their yield composition polydispersity at target bandgaps, spanning 1.9 2.9 eV, simultaneously tuned. Utilizing eleven precision‐tailored QD compositions obtained without any prior knowledge, within 30 h, using less than 210 mL total starting solutions, user experiments. Using knowledge generated from studies, pre‐trained use a new batch precursors further accelerate synthetic path discovery compositions, by least twofold. knowledge‐transfer strategy enhances optoelectronic properties in‐flow synthesized QDs (within same resources as no‐prior‐knowledge experiments) mitigates issues batch‐to‐batch precursor variability, resulting averaging 1 meV peak emission energy.

Язык: Английский

---

Autonomous experimentation systems for materials development: A community perspective

Matter, Год журнала: 2021, Номер 4(9), С. 2702 - 2726

Опубликована: Июль 26, 2021

Язык: Английский

---

Autonomous chemical research with large language models

Nature, Год журнала: 2023, Номер 624(7992), С. 570 - 578

Опубликована: Дек. 20, 2023

Transformer-based large language models are making significant strides in various fields, such as natural processing

Язык: Английский

---

Emerging materials intelligence ecosystems propelled by machine learning

Nature Reviews Materials, Год журнала: 2020, Номер 6(8), С. 655 - 678

Опубликована: Ноя. 9, 2020

Язык: Английский

---

AiZynthFinder: a fast, robust and flexible open-source software for retrosynthetic planning

Journal of Cheminformatics, Год журнала: 2020, Номер 12(1)

Опубликована: Ноя. 17, 2020

We present the open-source AiZynthFinder software that can be readily used in retrosynthetic planning. The algorithm is based on a Monte Carlo tree search recursively breaks down molecule to purchasable precursors. guided by an artificial neural network policy suggests possible precursors utilizing library of known reaction templates. fast and typically find solution less than 10 s perform complete 1 min. Moreover, development code was range engineering principles such as automatic testing, system design continuous integration leading robust with high maintainability. Finally, well documented make it suitable for beginners. available at http://www.github.com/MolecularAI/aizynthfinder .

Язык: Английский

---

Machine learning the ropes: principles, applications and directions in synthetic chemistry

Chemical Society Reviews, Год журнала: 2020, Номер 49(17), С. 6154 - 6168

Опубликована: Янв. 1, 2020

Chemists go ML! This tutorial review provides easy access to the fundamentals of machine learning from a synthetic chemist's perspective. Its diverse applications for molecular design, synthesis planning, or reactivity prediction are summarized.

Язык: Английский

---

Artificial intelligence in drug discovery: what is realistic, what are illusions? Part 1: Ways to make an impact, and why we are not there yet

Drug Discovery Today, Год журнала: 2020, Номер 26(2), С. 511 - 524

Опубликована: Дек. 18, 2020

Although artificial intelligence (AI) has had a profound impact on areas such as image recognition, comparable advances in drug discovery are rare. This article quantifies the stages of which improvements time taken, success rate or affordability will have most overall bringing new drugs to market. Changes clinical rates improving discovery; other words, quality decisions regarding compound take forward (and how conduct trials) more important than speed cost. current AI focus make given compound, question make, using efficacy and safety-related end points, received significantly less attention. As consequence, proxy measures available data cannot fully utilize potential discovery, particular when it comes safety vivo. Thus, addressing questions generate points model be key clinically relevant decision-making future.

Язык: Английский

---