GROQ-seq: A Collaborative, Open Data Approach to Addressing Protein Function Prediction

David Ross, Aviv Spinner, Simon d'Oelsnitz, Svetlana P Ikonomova, Olga Vasilyeva, Nina Alperovich, Kristen Sheldon, Courtney Tretheway, Dana Cortade, Erika DeBenedictis, Peter J Kelly

Track: Biology: datasets and/or experimental results

Nature Biotechnology: Yes

Keywords: protein function, ML, machine learning, data

TL;DR: We have developed an experimental platform and unified data ontology for collecting and sharing open-access data on diverse protein functions to enable the development of predictive models linking sequence to function

Abstract: We have developed an experimental platform and unified data ontology for collecting and sharing open-access data on diverse protein functions to enable the development of predictive models linking sequence to function. The experimental strategy for data generation employs the growth-based quantitative sequencing (GROQ-seq) platform as a simple yet adaptable system that can be easily expanded to encompass new functions. This high-throughput experimental platform can produce quantitative functional characterization data for hundreds of thousands of proteins per experiment at a cost of approximately $0.05 per sequence. To date, we have made significant progress in collecting data for our initial protein function: transcription factor binding. We are also developing GROQ-seq for a suite of additional protein functions, including proteases, aminoacyl tRNA synthetases, RNA polymerases, histidine kinases, single-chain antibody fragments, and a variety of metabolic enzymes. Being both highly-scalable and extensible to new protein functions, the GROQ-seq platform enables the collection of the diversity of data necessary to create a generalizable model that quantitatively predicts sequence to function relationships.

Anonymization: This submission has been anonymized for double-blind review via the removal of identifying information such as names, affiliations, and identifying URLs.

Presenter: ~Aviv_Spinner1

Format: Yes, the presenting author will attend in person if this work is accepted to the workshop.

Funding: No, the presenting author of this submission does *not* fall under ICLR’s funding aims, or has sufficient alternate funding.

Submission Number: 85