Reading genomes, bit by bit
Because of rapid advances in genome sequencing technology, we can finally see the source code for life: the complete genomic DNA sequences that specify development, regulation, and function of organisms. We still don't really understand how to read this trove of encoded information, and we don't really understand in any satisfying detail how it evolved.
Analogies to reading, source code, and encoding quickly break down. Genomes are digital information, and can be analyzed with tools that have parallels in digital signal processing fields, but genomes work very differently from human digital technologies. Using computational analysis to figure out how genomes work and how they evolved is a whole field of its own: computational genomics.
Our laboratory develops computational methods for genome sequence analysis. We are particularly interested in methods for identifying remote evolutionary relationships between distantly related protein and RNA sequences.
We're a Howard Hughes Medical Institute laboratory at Harvard University, in the Molecular & Cellular Biology Department. We also have affiliations to Applied Mathematics, the FAS Center for Systems Biology, the Center for Brain Science, and to the Broad Institute.
Recent publications
- Evolution: Getting Something for Nothing.
C. M. Weisman, S. R. Eddy.
Current Biology, in press, 2017. -
Inferring Functional Neural Connectivity with Deep Residual Convolutional Networks.
T. W. Dunn, P. K. Koo.
bioRxiv, 2017.
preprint
How to reach us
Department of Molecular & Cellular BiologyBiological Laboratories 1008A
16 Divinity Avenue
Harvard University
Cambridge MA 02138, USA
HMMER
Hidden Markov models for sequence profile analysis.
Infernal
RNA structure analysis using covariance models.
Databases
Pfam
Database of protein family alignments and hidden Markov models.
Rfam
The Rfam database of RNA alignments, consensus secondary structures, and profile SCFGs.
Dfam
The Dfam database of repetitive DNA sequence elements.