Software for determining the 3D structure of proteins

Written by
Catherine Zandonella, Office of the Dean for Research
Nov. 3, 2020

A new software package aims to aid drug design and biomedical research by making it easy to construct 3D images of proteins and other molecules using one of the world’s most powerful microscopes.

Amit Singer, professor of mathematics and the Program in Applied and Computational Mathematics, and his team are developing a package they call Algorithms for Single Particle Reconstruction, or ASPIRE, that takes in 2D images captured by cryo-electron microscopy and produces reliable 3D structures without significant human intervention.

The package will offer fully automated and faster data processing, producing highly accurate images. Whereas existing software packages require human input on which images to include in analysis, ASPIRE needs little user modification, reducing the potential for bias. Unlike today’s software, which starts with an initial estimate of the 3D structure, ASPIRE does not make any initial assumptions but rather relies only on the images themselves. The team’s long-term goal is to develop a commercial software package that will make biomolecular structures more readily available for drug discovery and research.

Singer’s research was featured at the annual Celebrate Princeton Innovation (CPI) event in November 2019 that highlights the work of faculty and student researchers who are making discoveries and creating inventions with the potential for having broad societal impact.

This year, CPI is part of Engage 2020, a multi-day virtual conference being held Nov. 4-6 to create new connections among Princeton innovators and leaders in entrepreneurship, industry, nonprofit organizations, and government in the state, regional and global innovation ecosystems. The CPI portion of the program is 3:30-6 p.m. Thursday, Nov. 5.

View main article.