Data Science Investment Counter —Funding raised by UK data science companies in 2018.
£ 5.640 Million

New machine learning technique analyses nanomedicines for cancer immunotherapy

SNAs are ball-like forms of DNA and RNA arranged on the surface of a nanoparticle.
Credit: Chad Mirkin/Northwestern University

Scientists from Northwestern University have developed a direct route to optimise spherical nucleic acids (SNAs), bringing them one step closer to becoming a viable treatment option for many forms of cancer, genetic diseases, neurological disorders and more.

SNAs are a class of personalised medicines. They are challenging to optimise because their structures can vary in many ways, but the team led by nanotechnology expert Chad A. Mirkin developed a library approach and machine learning to optimally synthesise, analyse and select these nanostructures rapidly.

The related study that was published this week in Nature Biomedical Engineering details the optimisation method, mentioning how more than 1,000 structure were analysed by the system in the process.

“Spherical nucleic acids represent an exciting new class of medicines that are already in five human clinical trials for treating diseases, including glioblastoma (the most common and deadly form of brain cancer) and psoriasis,” said Mirkin, the inventor of SNAs and the George B. Rathmann Professor of Chemistry in Northwestern’s Weinberg College of Arts and Sciences.

“This study shows that we can address the complexity of the SNA design space, allowing us to focus on and exploit the most promising structural features of SNAs, and ultimately, to develop powerful cancer treatments“, he added.

According to Mirkin’s team, the research revealed that SNAs variation in structure leads to biological activities affecting the efficacy of these nanostructures. Because these relationships were not predicted, they likely would have gone unnoticed in a typical study of a small set of structures.

For instance, the ability to stimulate an immune response can depend on nanoparticle size or their composition. “With this new information, researchers can rank the structural variables in order of importance and efficacy, and help establish design rules for SNA effectiveness,” said Andrew Lee, assistant professor of chemical and biological engineering in the McCormick School of Engineering and study co-author.

The Nature Biomedical Engineering paper about the new research is named “Addressing Nanomedicine Complexity with Novel High-Throughput Screening and Machine Learning.” The whole text is available on the journal’s website at this link.


Co-working space and blog dedicated to all things data science.

Subscribe to our newsletter