Eric D. Siggia, Ph.D.
Professor
Laboratory of Theoretical Condensed Matter Physics
E-mail: siggiae@rockefeller.edu
As large amounts of genetic data, such as genome sequences, have been generated, the field of bioinformatics has become
essential for processing and analyzing biological information. Dr. Siggia’s laboratory uses the bioinformatics approach
to study gene control in bacteria, yeast and flies, inventing probability-based models to discern the regulatory patterns involved in gene expression.
The genome is more than a parts list — it functions
as an assembly manual that directs when
and where genes are to be expressed in response to
signals conveyed by proteins. Although new technologies
have been perfected that provide a
genome-wide view of the expression of mRNA,
the blueprint for proteins, it has still not become
clear how regulatory DNA codes for mRNA’s
expression. Decoding the regulatory part of the
genome is challenging, however, as more genomes
are sequenced, it becomes possible to compare regions
of regulatory DNA and infer functional units.
Dr. Siggia’s lab has developed an algorithm
called Stubb to identify regulatory modules computationally.
Using Stubb, his lab has scanned
the genome of the fruit fly Drosophila melanogaster
to calculate the number and location of
clusters of binding sites. Fruitful tests include the
segmentation gene hierarchy, which patterns the
developing embryo through a genetic cascade.
Dr. Siggia’s lab is using this signaling pathway as
a prototype of combinatorial control to determine
new blastoderm patterned genes as well as
new regulatory modules. The Stubb algorithm
also can examine related genomes and score
regulatory regions by their degree of functional
variation. By computationally screening the large
set of blastoderm modules, Dr. Siggia has found
modules that disappeared or moved to nonhomologous
but adjacent regions of the genome.
The Siggia lab has also exploited an underappreciated
resource to decipher the function of
regulatory DNA: modeling the binding preferences
of transcription factors by homology with
the database of protein-DNA co-crystals. The
lab has shown that even a modest bias toward
the correct motif can substantially boost the
performance of tools that rely only on sequence.
Dr. Siggia and his colleagues have also modeled
the sequence dependence of the binding between
a nucleosome and DNA, thus looking simultaneously
at nucleosomes and factors as they compete
for binding to the DNA.
As the cost of resequencing entire genomes
decreases, new kinds of bioinformatics experiments
have become possible. In collaboration
with the Tomasz lab, Dr. Siggia has examined the
evolution of a Staphylococcus aureus infection
in a single patient as it developed resistance to
antibiotics over a three month period. The Siggia
lab sequenced the first and last isolates and
found only 34 point mutations, which appeared
incrementally in the strains isolated at intermediate
times from the patient. By analyzing the
sequences, Dr. Siggia identified about a quarter
that could plausibly impact resistance, and
showed that a number of the same operons were
also mutated in other resistant strains.
Another collaboration, with Rockefeller’s
Frederick Cross, uses analysis of gene expression
in yeast to design experiments that probe the
“grammar” of regulatory elements. Though
seldom stated, most of the sites of the best-characterized
factors in yeast do not imply expression
for their cognate genes, and chip analysis and
comparative sequence data has not yet revealed
what the inhibitory motifs might be. Selection
experiments in yeast, using randomly ligated
regulatory elements, has generated hundreds of
functional and nonfunctional promoters, built
from a very limited number of “words.”
Another area of interest for the Siggia lab is in
modeling cell division. The cell cycle in budding
yeast is a network for which most of the parts
are known, yet predicting how they will work
together remains a challenge. Together with the
Cross lab, Dr. Siggia is making movies of growing
yeast colonies over five or more divisions while
following various markers. The analysis of these
movies using custom software has led to several
new insights into the transitions between phases
of the cell cycle, the effects of various feedbacks,
and the function of “well studied” genes. The
natural fluctuations of the cell cycle provide an
assay for size control, unbiased by gene deletion,
which Dr. Siggia suggests will lead to a quantitative
model of this illusive phenomenon. Using a
new flow cell technology, he and his colleagues
have recently worked to phase lock the cell cycle
to an external clock in much the same way that
circadian oscillators are locked in phase to light.
CAREER
Dr. Siggia received his A.B. in physics in 1971
and his Ph.D. in physics in 1972, both from
Harvard University. He was a junior fellow
at Harvard from 1972 to 1975, and then was
assistant professor at the University of Pennsylvania
for two years before moving to Cornell
University where he became professor of physics
in 1985. He has been a long-term visitor or
consultant at the University of Paris in Orsay,
École Normale Supérieure in Paris, the Santa
Barbara Institute for Theoretical Physics, Bell
labs and Los Alamos National Laboratory. Dr.
Siggia came to Rockefeller University in 1997.
Dr. Siggia received a John Simon Guggenheim
Fellowship in 1988. He was an Alfred P.
Sloan Research Fellow from 1980 to 1982.
