AI Models for Future Biology Research

A new model by MIT researchers significantly eases the difficulty of integrating AI into biology research

Ontario Youth Medical Society
3 min readNov 25, 2023
Image by author.

The field of biological research and healthcare is undergoing a seismic shift, thanks to a groundbreaking innovation by MIT scientists — BioAutoMATED. This new AI tool was designed in hopes of making biological research less costly and time-consuming for all scientists, and to accelerate the world of scientific research.

This tool was created by Jim Collins, the Termeer Professor of Medical Engineering and Science in the Department of Biological Engineering at MIT and the life sciences faculty lead at the Abdul Latif Jameel Clinic for Machine Learning in Health, alongside countless other colleagues.

Their aim for this tool was to solve the issues that typically come up in machine learning research — the cost of time, money, and work that it takes to select a fitting model and format the dataset for said model. Each of these factors plays a role in how the model performs, which can in turn effect the research as a whole. Another reason that this team decided to assemble and create this incredible technology was that current automated machine-learning systems were still in the stages of development incapable of being used in fields such as biology.

Thus, as said by Luis Soenksen, a co-author and Jameel Clinic Postdoctoral Fellow, “Biological sequences such as DNA, RNA, proteins, and glycans have the amazing informational property of being intrinsically standardized, like an alphabet. A lot of AutoML tools are developed for text, so it made sense to extend it to [biological] sequences.”

At its core, BioAutoMATED is an incredible automation system specifically crafted to redefine how biological research is conducted. It’s not just about efficiency; it’s a catalyst for future advancements, being able to conduct complex experiments and handling complicated data analysis.

The BioAutoMated tool can currently be used with three types of models: binary classification models (dividing data into two classes), multi-class classification models (dividing data into multiple classes), and regression models (fitting continuous numerical values or measuring the strength of relationships of variables). The code for this technology is open and available to all researchers, and feedback is telling the team that it’s convenient and easy to run.

“What we would love to see is for people to take our code, improve it, and collaborate with larger communities to make it a tool for all,” Luis Soenksen says in an MIT News article.

Image by BioAutoMATED: An end-to-end automated machine learning tool for explanation and design of biological sequences research paper.

If you wish to learn more about this new technology, the scientists have written an open-access paper on their proposed solution to the current inefficiency of machine-learning models. Click here to learn more about BioAutoMATED!

In conclusion, MIT’s BioAutoMATED is not just a technological tool; it’s a revolutionary technology that can improve the world of science and even healthcare. Its use has the potential to change the approach to biological research, which in turn can accelerate healthcare projects such as finding cures for cancer or treatments for genetic conditions. As scary as AI can be in the world right now, in the right hands of scientists and healthcare workers, its use by these experts can be a benefit!

Sources

Valeri, J. A. et al. (2023, June 21). BioAutoMATED: An end-to-end automated machine learning … — cell press. Cell Systems. https://www.cell.com/cell-systems/fulltext/S2405-4712(23)00151-5

MIT scientists build a system that can generate AI models for biology researchBioAutoMATED is a new MIT system that can generate artificial intelligence models for biology research. The…news.mit.edu

About the Author

Jenny Pham is a McMaster Life Sci student. She is currently working as a Math & English tutor, a volunteer with Peel Health at a long-term care center, a passionate MEDLIFE chapter founder, and an amateur writer looking to combine her passions for medicine and writing. Like others around her, she’s a second-generation immigrant working towards a career in medicine and education, looking to continue advocating for underprivileged communities as a physician, through research, and education. You can connect with her on LinkedIn!

--

--

Ontario Youth Medical Society
Ontario Youth Medical Society

Written by Ontario Youth Medical Society

Ontario Youth Medical Society is a student-led, non-profit organization focused on educating youth and making a difference in medicine.

No responses yet