Source – genengnews.com
Streamlining Academic Research
Insilico Medicine, a cutting-edge biotechnology firm known for its innovations in AI-driven drug discovery, has unveiled Science42, a new tool designed to streamline the academic writing process. Academic writing is essential for the dissemination of scientific knowledge, but it often poses significant challenges, particularly for early-career researchers and non-native English speakers. According to a 2018 survey published in Nature, nearly 37% of researchers reported dedicating over 20 hours per week to writing and revising scientific papers, highlighting the considerable time investment required.
Harnessing AI for Improved Drafting
The launch of Science42 marks a significant advancement in addressing these challenges. This new draft outline research assistant integrates advanced Natural Language Processing (NLP) technology, leveraging Generative Pre-trained Transformers (GPT) and other Large Language Models (LLMs). These tools enable researchers to process and generate extensive amounts of literature with greater efficiency. Science42 is designed to assist in drafting a wide range of scientific documents, including research papers, grant applications, patent filings, internal research summaries, and Investigational New Drug (IND) applications.
Petrina Kamya, PhD, Global Head of AI Platform and Vice President at Insilico Medicine, explained the motivation behind Science42. “One of the most challenging aspects of writing is getting started. This was a personal experience of mine during my time as a graduate student when I had to write numerous grants, papers, and reports,” said Kamya. “Science42 was developed to remove that initial barrier and make the writing process smoother and more efficient.”
Collaborative Validation and Future Prospects
To validate the capabilities of Science42, Insilico Medicine collaborated with researchers from the University of Copenhagen to draft a paper on medRxiv. The study, which compares radiotherapy outcomes for different brain tumor types, including Glioblastoma Multiforme and Low-Grade Gliomas, utilized data from 32 cancer datasets. While the paper was initially drafted by DORA, it underwent manual curation and extension to ensure accuracy and completeness.
Insilico Medicine plans to further test Science42 across various document types and is preparing to release a free trial version to the public in late 2024. This initiative reflects Insilico’s broader mission to leverage generative AI for drug discovery and development. The company has been at the forefront of applying generative AI to create novel molecules since 2016, and its AI-driven Pharma.AI platform integrates advancements in biology, chemistry, and medicine.
Since 2021, Insilico Medicine has identified 18 preclinical candidates from its portfolio of over 30 therapeutic assets, advancing seven of these molecules to clinical trials. Notably, a recent paper published in Nature Biotechnology details the experimental data and evaluations of Insilico’s lead drug, a TNIK inhibitor for idiopathic pulmonary fibrosis, currently undergoing Phase II trials.
With the introduction of Science42, Insilico Medicine aims to enhance the efficiency of academic writing and support researchers in producing high-quality scientific documents with greater ease.
