What question did this study set out to answer?

To enhance information density in ML-driven science via innovative wet lab methods and learning algorithms.

December 22, 20250 citationsOpen Access

Lifting Biomolecular Data Acquisition

EWEli N. WeinsteinBioFactura (United States)ASAndrei SlabodkinOslo University Hospital MGMattia G. GollubBioFactura (United States)

Key Points

To enhance information density in ML-driven science via innovative wet lab methods and learning algorithms.
Developed a method using neural extension of compressed sensing.
Conducted simultaneous measurements of multiple molecules.
Implemented a co-design approach for experiments and algorithms.
Achieved significant gains in information density.
Demonstrated effectiveness using antibodies and cell therapies.

Abstract

One strategy to scale up ML-driven science is to increase wet lab experiments' information density. We present a method based on a neural extension of compressed sensing to function space. We measure the activity of multiple different molecules simultaneously, rather than individually. Then, we deconvolute the molecule-activity map during model training. Co-design of wet lab experiments and learning algorithms provably leads to orders-of-magnitude gains in information density. We demonstrate on antibodies and cell therapies.

Ask AI

Helpful

Bookmark

View Full Paper