Improving Inverse Folding models at Protein Stability Prediction without additional Training or Data
Dutton, O.; Bottaro, S.; Invernizzi, M.; Redl, I.; Chung, A.; Hoffmann, F.; Henderson, L.; Ruschetta, S.; Airoldi, F.; Owens, B. M.; Foerch, P.; Fisicaro, C.; Tamiola, K.
Show abstract
Deep learning protein sequence models have shown outstanding performance at de novo protein design and variant effect prediction. We substantially improve performance without further training or use of additional experimental data by introducing a second term derived from the models themselves which align outputs for the task of stability prediction. On a task to predict variants which increase protein stability the absolute success probabilities of PO_SCPLOWROTEINC_SCPLOWMPNN and ESMO_SCPLOWIFC_SCPLOW are improved by 11% and 5% respectively. We term these models PO_SCPLOWROTEINC_SCPLOWMPNN-O_SCPLOWDDC_SCPLOWG and ESMO_SCPLOWIFC_SCPLOW-O_SCPLOWDDC_SCPLOWG.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
Similar papers in this journal
Similar papers in this journal
- AlphaMut: a deep reinforcement learning model to suggest helix-disrupting mutations 95%
- Enzyme stability-activity trade-off: new insights from protein stability weaknesses and evolutionary conservation 95%
- Thermal Adaptation of Cytosolic Malate Dehydrogenase Revealed by Deep Learning and Coevolutionary Analysis 94%
"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.