Pytorch model for predicting solvent accessible surface area of Fvs.
Inspired by RaptorX-Contact/Property, AlphaFold1 and trRosetta, I wanted to build a model that predicted a distribution of SASAs for Fabs. Downstream, this can be useful in predicting solubility/aggregation potential of mAbs.
Here's a more detailed writeup on the initial results of this project: https://r-krishna.github.io/blog/2021/05/27/resnet-antibody-surface
Citations:
Jeffrey A Ruffolo, Carlos Guerra, Sai Pooja Mahajan, Jeremias Sulam, Jeffrey J Gray, Geometric potentials from deep learning improve prediction of CDR H3 loop structures, Bioinformatics, Volume 36, Issue Supplement_1, July 2020, Pages i268–i275, https://doi.org/10.1093/bioinformatics/btaa457
S. Wang, S. Sun, Z. Li, R. Zhang and J. Xu, "Accurate De Novo Prediction of Protein Contact Map by Ultra-Deep Learning Model", PLOS Computational Biology, vol. 13, no. 1, p. e1005324, 2017. Available: 10.1371/journal.pcbi.1005324.
J. Yang, I. Anishchenko, H. Park, Z. Peng, S. Ovchinnikov and D. Baker, “Improved protein structure prediction using predicted interresidue orientations.,” Proceedings of the National Academy of Sciences, 2020. PNAS