PPubMed14 Aug 2024Cited 12×

Generative AI in drug discovery and development: the next revolution of drug discovery and development would be directed by generative AI

Chiranjib Chakraborty, Manojit Bhattacharya, Soumen Pal, Md. Aminul Islam

Abstract

'Md. Aminul Islam'] Dear Editor, Previously, we published a correspondence article in International Journal of Surgery about next-generation drug discovery and development using ChatGPT or Large Language Model (LLM)1. The article was very timely. However, we found that generative artificial intelligence (AI) is progressing quickly in drug discovery and development. In this article, we discuss the role of generative AI in drug discovery and development.

A figure from Generative AI in drug discovery and development: the next revolution of drug discovery and development would be directed by generative AI — fig. from the paper

§ The Valyu brief

The paper, in five minutes

Read from the full text via Valyu Contents

Generative AI is rapidly transforming drug discovery, cutting costs and time while enabling new drug candidates to reach clinical trials.

What’s new

i.The article reviews recent generative AI tools (AlphaFold, ChemSpaceAL, GraphGPT, PETrans, DrugChat) used in drug discovery.
ii.It reports that about 15 AI‑designed drug candidates have entered clinical trials, including REC‑2282 and BEN‑8744.
iii.It quantifies potential cost and time reductions of 25‑50% from AI‑driven approaches.
iv.It outlines the scale of large language models (GPT‑3, GPT‑4) and their relevance to drug design.
v.It discusses current limitations such as data dependence, validation needs, and reproducibility issues.

How they did it

The authors performed a literature‑based review, summarizing recent studies, tools, and clinical examples of generative AI in drug discovery and development.

What they found

AI‑designed drugs REC‑2282 (Phase 2/3) and BEN‑8744 (Phase 1) have entered clinical trials.
Approximately 15 AI‑generated drug candidates have reached various clinical trial stages.
AI is estimated to reduce drug discovery time and cost by 25‑50%.
GPT‑4 was trained on ~13 trillion tokens and 1.8 trillion parameters, enabling advanced molecule design.

Caveats

✕The accuracy of generative AI models depends heavily on the quality of training datasets.
✕Models require extensive validation and testing to confirm reliability.
✕The reproducibility crisis affects target identification and model predictions.
✕AlphaFold predicts only a single protein state, limiting dynamic insights.
✕LLMs can produce misleading information and errors.

§ The Valyu brief

Reading the full paper and taking notes. This takes a few seconds…

§ Ask this paper

Ask a question about this paper

Valyu reads the full text and answers from what the paper actually says.

§ Related work in other archives

Similar papers in all four archives

The same idea searched in arXiv, PubMed, bioRxiv and chemRxiv at once, so a method can lead you to where it is being applied.

AarXiv4 found

PPubMed3 found

BbioRxiv4 found

CchemRxiv3 found

Searching the other archives…