From Spectra to Structure: An Interview with Novogaia’s Tess Bevers - Summary - MDSpire
Feature

From Spectra to Structure: An Interview with Novogaia’s Tess Bevers

Novogaia’s transformer-based model converts untargeted mass spectra into structural predictions for early-stage drug discovery

Share

Gaia-01, a machine-learning model developed by Novogaia, is revolutionizing natural product drug discovery by predicting molecular structures from mass spectrometry data. This AI-enhanced approach streamlines the identification of bioactive compounds, particularly from fungi, by transforming complex mass spectrometry signals into actionable insights. The model, evaluated using the MassSpecGym framework, demonstrates a significant performance leap, enhancing drug development pipelines while addressing challenges like data volume and benchmarking in computational metabolomics.

Original Source(s)

From Spectra to Structure: An Interview with Novogaia’s Tess Bevers

  • by Henry Thomas

  • April 15, 2026

Related Content

Mass Spec Roundup: From Masterpieces to Microplastics

Advances in imaging, AI, and single-cell analysis reveal hidden chemistry in art, biology, and the environment

  • April 15, 2026

  • 6 min
Cell and Gene Roundup: CAR T Breaks New Ground

A campus-built therapy enters clinical trials as researchers refine gene editing and uncover new ways to control gene expression

  • April 15, 2026

  • 6 min
Cancer Mutation Detection with Digital PCR

Evaluating rapid assays for the detection of somatic mutations in tissue and liquid biopsies

  • April 15, 2026

  • 9 min