Contact us     |     Newsletter subscription     |     Code of Conduct

ABOUT DEBORA MARKS

Debora established her laboratory 5 years ago after a career in industry and more recent degrees in mathematics and computational biology, aiming to accelerate fundamental discoveries in biomedicine. Developing robust statistical methods including unsupervised machine learning, Debora’s lab was able predict 3-dimensional protein structures from sequence alone, predict the fitness effects of human genetic variation, make robust generative models for therapeutic and antibody design and design deimmunization. Most recently she has extended these new tools to apply in dimension reduction and multimodal modelling of diverse biological and clinical data, including benchmarked approaches to combinations of RNA, protein and image data.

Mission to develop AI for design of biological interventions for human health and social justice.

Prediction and design of biological sequences with neural machines

Monday, 7 September | 4:30 pm (CEST)

What can we do with a million or a billion genomes? Understanding how variation across genomes shapes the properties of biomolecules, cells, and organisms is a foundational question in biology. I will demonstrate how probabilistic generative modeling of genetic variation can give surprisingly direct answers to questions about 3D structures, dynamics, the effects of mutations and the design of biological systems. Our new work extends from the undirected models of genetic variation to deep directed models using newly developed variational autoencoders, and autoregressive models that do not depend on sequence alignments. From purely unsupervised learning, we have improved of prior-art for predicting the effects of mutations and successfully designed optimized antibody libraries. I will introduce challenges for extending these methods to diverse biomedical and engineering applications, with specific examples of successful probabilistic models to generate novel functional biotherapeutics.

https://marks.hms.harvard.edu/ 
https://github.com/debbiemarkslab