Back

PATCRdb: Database of TCRs from data mining patent documents

Lee, Y.; Freitag, R.; Ganesan, R.; Schwammle, V.; Kumar, S.; Krawczyk, K.

2023-01-07 allergy and immunology
10.1101/2023.01.05.23284150 medRxiv
Show abstract

T-cells are crucial actuators of the innate immune system. Because their receptors recognize intracellular disease markers, there is considerable interest in developing them as novel biotherapies. Computational methods to support discovery, design and development of TCR-based therapeutics need robust repositories of curated sequence and structural information on TCRs. The urgency of this need is highlighted by the recent approval of the first TCR biotherapeutic, tebentafusp. In this work, we have collected patent data on TCR sequences to provide early access to TCRs that are in various stages of product and clinical development (pre-FDA approvals) and are already past the initial discovery / proof of concept (scientific publications) stages. We employ literature mining to identify patent documents disclosing TCR sequences. Such documents are further analyzed to provide a birds-eye view of TCR patenting landscape. We compile the information into a database available at http://github.com/konradkrawczyk/patcrdb that we hope should help TCR engineers.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.