Presenter: Vladimir Gligorijevic, Ph.D.,
Research Scientist, Systems Biology Group, Center for Computational Biology
Topic: Machine learning-guided protein design
Computational protein design seeks to create a novel protein sequence that folds to a given structure or carries out a given function. Self-supervised representation learning approaches offer the potential to explore the protein sequence and structure space more effectively and accelerate protein engineering. In this talk, I will discuss our novel self-supervised ML models for sequence-based and structure-based design of proteins with specific functions. I will introduce a novel architecture that combines graph denoising autoencoders (GDAE) with an external function analyser. GDAE, composed of a graph convolutional encoder with a self-attention decoder, learns the low-dimensional manifold of protein structures. GDAE is used for sampling novel protein structures consistent with the local manifold structure that are further optimized for desired functions using a pre-trained deepFRI model as the function analyser.
Join from PC, Mac, Linux, iOS or Android: https://simonsfoundation.zoom.us/j/286758074
Or Telephone:
Dial(for higher quality, dial a number based on your current location):
US: +1 312 626 6799 or +1 646 558 8656 or +1 346 248 7799 or +1 669 900 6833 or +1 253 215 8782 or +1 301 715 8592
Meeting ID: 286 758 074
International numbers available: https://simonsfoundation.zoom.us/u/abSaUoPIqD
Or an H.323 room system:
162.255.36.11 (US East)
Meeting ID: 286 758 074