Research – Statistical Inference & Machine Learning

PdirWe are interested in developing new algorithms and tools to solve biological problems inspired by the abundance of biological information generated by current technologies. Our lab is interested in applying and developing new  principles of information theory, coding theory, statistical physics, statistical estimation and machine learning to understand biology from a data driven approach. Another important objective of our lab is to create computational physical models of biomolecules and systems that can help us unravel fundamental biological phenomena and produce testable hypothesis to improve our understanding of molecular processes and disease related mechanisms. Please find below some representative publications related to this area of interest in our lab:

Jonathan Martin, Marcos Lequerica Mateos, José N Onuchic, Ivan Coluzza, Faruck Morcos. Machine learning in biological physics: From biomolecular prediction to design. Proceedings of the National Academy of Sciences. 121(27): e2311807121, 2024.


[PDF]

Cheyenne Ziegler, Jonathan Martin, Claude Sinner, Faruck Morcos. Latent generative landscapes as maps of functional diversity in protein sequence space. Nature Communications. 14(1): 2222, 2023.


[PDF]

Krithika Ravishankar, Xianli Jiang, Emmett M Leddin, Faruck Morcos, G Andrés Cisneros. Computational compensatory mutation discovery approach: Predicting a PARP1 variant rescue mutation. Biophysical Journal. 121(19): 3663-3673, 2022.


[PDF]

Jeffrey K. Noel, Faruck Morcos and Jose N. Onuchic. Sequence co-evolutionary information is a natural partner to minimally-frustrated models of biomolecular dynamics F1000Research 5(F1000 Faculty Rev):106. 5:13652, 2016.


[PDF]

Ryan Cheng, Faruck Morcos, Herbert Levine and Jose N. Onuchic. Towards rationally redesigning bacterial two-component signaling systems using coevolutionary information. Proc Natl Acad Sci USA . 111(5): E563-71, 2014
Press Highlight in Phys.org: Researchers tune in to protein pairs: Team quanti es how mutations a ffect cell signaling in bacteria


[PDF]

Faruck Morcos, Biman Jana, Terence Hwa and Jose N. Onuchic. Coevolutionary signals across protein lineages help capture multiple protein conformations. Proc Natl Acad Sci USA. Vol 110, No. 51, p. 20533-20538. 2013
Press Highlight in ScienceDaily: Proteins' passing phases revealed

 
[PDF]
 

Ramy Mourad, Zaher Dawy and Faruck Morcos. Designing Pooling Systems for Noisy High-Throughput Protein-Protein Interaction Experiments using Boolean Compressed Sensing. IEEE/ACM Transactions on Computational Biology and Bioinformatics. 10(6): 1478-1490, 2013.

 
[PDF]
 

Joanna I. Sulkowska*, Faruck Morcos*, Martin Weigt, Terence Hwa and Jose N. Onuchic. Genomics-aided structure prediction. Proc Natl Acad Sci USA. Vol. 109, No. 26, p. 10340-45, 2012. (*equivalent contribution)
Protein residues kiss, don't tell: Genomes reveal contacts, scientists refine methods for protein-folding prediction

 
[PDF]
 

Faruck Morcos*, Andrea Pagnani*, Bryan Lunt, Arianna Bertolino, Debora S. Marks, Chris Sander, Riccardo Zecchina,
Jose N. Onuchic, Terence Hwa and Martin Weigt. Direct-coupling analysis of residue coevolution captures native contacts across many protein families. Proc Nat Acad Sci USA. Vol. 108, No. 49, E1293-E1301, 2011. (*equivalent contribution)

 
[PDF]
 

Faruck Morcos, Marcin Sikora, Mark Alber, Dale Kaiser and Jesus A. Izaguirre. Belief Propagation Estimation of Protein and Domain Interactions using the Sum-Product Algorithm. IEEE Transactions on Information Theory. Vol. 56, No. 2, p. 742-755, 2010.

 
[PDF]