Timezone: »

 
Probabilistic thermal stability prediction through sparsity promoting transformer representation
Yevgen Zainchkovskyy · Jesper Ferkinghoff-Borg · Anja Bennett · Thomas Egebjerg · Nikolai Lorenzen · Per Greisen · Søren Hauberg · Carsten Stahlhut

@
in Robustness in Sequence Modeling »
Event URL: https://openreview.net/forum?id=ZUHPiSbvoX »

Pre-trained protein language models have demonstrated significant applicability in different protein engineering task. A general usage of these pre-trained transformer models latent representation is to use a mean pool across residue positions to reduce the feature dimensions to further downstream tasks such as predicting bio-physics properties or other functional behaviours. In this paper we provide a two-fold contribution to machine learning (ML) driven drug design. Firstly, we demonstrate the power of sparsity by promoting penalization of pre- trained transformer models to secure more robust and accurate melting temperature (Tm) prediction of single-chain variable fragments with a mean absolute error of 0.23C. Secondly, we demonstrate the power of framing our prediction problem in a probabilistic framework. Specifically, we advocate for the need of adopting probabilistic frameworks especially in the context of ML driven drug design.

Author Information

Yevgen Zainchkovskyy (Technical University of Denmark)
Jesper Ferkinghoff-Borg (Novo Nordisk A/S)
Anja Bennett
Thomas Egebjerg (Novo Nordisk AS)
Nikolai Lorenzen
Per Greisen (Novo Nordisk)
Søren Hauberg (Technical University of Denmark)
Carsten Stahlhut (Novo Nordisk A/S)

More from the Same Authors