The Datasets

Living datasets backed by automated experiments and open source methods.

DATASET

Protein Sequence to Function

Protein functions, such as enzymatic activities, binding interactions, and membrane transport, exist as islands in the “archipelago” of the protein function landscape. Machine learning (ML) algorithms have tried to bridge this gap, but are still unable to find a general solution for predicting any protein’s function from its DNA sequence.

A general solution for predicting any protein function from sequence would catalyze a transformation in the field of biology.

We propose to develop an experimental platform and unified data ontology for collecting datasets from different functional ‘islands’ to build predictive models for individual protein functions. The experimental strategy uses a pooled, growth-based assay measured with DNA sequencing to create a simple, yet adaptable system that can be easily expanded to encompass new functions.

Growth-Based Assays for Measuring Function (Full Proposal)

OUR STRATEGY

We view protein functions as islands in an archipelago. Our strategy is to align dataset creation across different protein function ‘islands’ to enable a generalized “sequence → function” predictive model.

Models developed from this data will initially succeed at predicting protein function within a single ‘island’, an individual family of proteins with a single function.

As the datasets grow and more islands are sampled, the models will become more generalized and capable of predicting the function of protein sequences that are increasingly distant from those that have been directly measured.

HOW ARE WE DOING IT?

We are utilizing massively pooled, growth-based assays enabling 100,000-500,000 data points per experiment at a cost of ~$0.05 per protein.

Growth-based assays link any protein-of-interest’s function with the ability for a cell containing it to grow.

The plasmid library of barcoded proteins is created and transformed into host cells. Then the cells are pooled, grown and challenged. The resulting cells are then sequenced, counting the number of barcodes present, which can be translated back into a measurement of quantitative function.
By simply changing elements of the plasmid’s gene circuit, the same methods and analyses can be used to interrogate new protein functions.

OUR TIMELINE

Our timeline for the first five protein function datasets.

Project Team

Dana Cortade

TECHNICAL PROJECT MANAGER

Team Member Spotlight

Dana Cortade holds a PhD from Stanford, where she focused on developing medical diagnostic devices. She has expertise in bioengineering, optics, and materials science. Dana’s experience across academic and nonprofit settings allows her to drive Align’s mission of enabling global collaboration through accessible technology platforms.

Proposal Contributors

David Ross—Living Measurement Systems Foundry, National Institute of Standards and Technology (NIST)

Simon d’Oelsnitz—Harvard Medical School, Harvard University

Erika DeBenedictis—Biodesign Lab, The Francis Crick Institute

Anjali Chadha—Biodesign Lab, The Francis Crick Institute

Oliver Hayes—Biodesign Lab, The Francis Crick Institute

Geoffrey Taghon—Living Measurement Systems Foundry, National Institute of Standards and Technology (NIST)

Mark Dörr—University of Greifswald Stefan Born - Technische Universität Berlin

Data Collection Partners