Connecting molecular properties with plain language

Glen M. Hocky

doi:10.1038/s42256-024-00812-y

Connecting molecular properties with plain language

Glen M. Hocky

Chemistry

Research output: Contribution to journal › Comment/debate › peer-review

Abstract

AI tools such as ChatGPT can provide responses to queries on any topic, but can such large language models accurately ‘write’ molecules as output to our specification? Results now show that models trained on general text can be tweaked with small amounts of chemical data to predict molecular properties, or to design molecules based on a target feature.

Original language	English (US)
Pages (from-to)	249-250
Number of pages	2
Journal	Nature Machine Intelligence
Volume	6
Issue number	3
DOIs	https://doi.org/10.1038/s42256-024-00812-y
State	Published - Mar 2024

ASJC Scopus subject areas

Software
Human-Computer Interaction
Computer Vision and Pattern Recognition
Computer Networks and Communications
Artificial Intelligence

Access to Document

10.1038/s42256-024-00812-y

Cite this

@article{70436e3e60e64247a5e24efb896f699a,

title = "Connecting molecular properties with plain language",

abstract = "AI tools such as ChatGPT can provide responses to queries on any topic, but can such large language models accurately {\textquoteleft}write{\textquoteright} molecules as output to our specification? Results now show that models trained on general text can be tweaked with small amounts of chemical data to predict molecular properties, or to design molecules based on a target feature.",

author = "Hocky, {Glen M.}",

note = "Publisher Copyright: {\textcopyright} Springer Nature Limited 2024.",

year = "2024",

month = mar,

doi = "10.1038/s42256-024-00812-y",

language = "English (US)",

volume = "6",

pages = "249--250",

journal = "Nature Machine Intelligence",

issn = "2522-5839",

publisher = "Springer International Publishing",

number = "3",

}

TY - JOUR

T1 - Connecting molecular properties with plain language

AU - Hocky, Glen M.

N1 - Publisher Copyright: © Springer Nature Limited 2024.

PY - 2024/3

Y1 - 2024/3

N2 - AI tools such as ChatGPT can provide responses to queries on any topic, but can such large language models accurately ‘write’ molecules as output to our specification? Results now show that models trained on general text can be tweaked with small amounts of chemical data to predict molecular properties, or to design molecules based on a target feature.

AB - AI tools such as ChatGPT can provide responses to queries on any topic, but can such large language models accurately ‘write’ molecules as output to our specification? Results now show that models trained on general text can be tweaked with small amounts of chemical data to predict molecular properties, or to design molecules based on a target feature.

UR - http://www.scopus.com/inward/record.url?scp=85187646941&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85187646941&partnerID=8YFLogxK

U2 - 10.1038/s42256-024-00812-y

DO - 10.1038/s42256-024-00812-y

M3 - Comment/debate

AN - SCOPUS:85187646941

SN - 2522-5839

VL - 6

SP - 249

EP - 250

JO - Nature Machine Intelligence

JF - Nature Machine Intelligence

IS - 3

ER -

Connecting molecular properties with plain language

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this