[CSBi-events] CSBi Seminar Series - Save the Date!

[csbi-events@mit.edu]

CSBi Community,

You may want to take note of the upcoming seminar on Friday, October 
21.  Thank you.


Dr. Philip Green
Howard Hughes Medical Institute and
Genome Sciences Department, University of Washington

Signal and Noise in Genome Sequences

Friday, October 21, 2005
Maclaurin Building (4-270)
3:00 - 4:00 p.m.
Light refreshments served.

Abstract
Interpreting genome sequences requires distinguishing 'signal' 
(encoded functional elements) from 'noise' (non-functional, neutrally 
evolving sequence). The characteristics of neutrally evolving 
sequence reflect underlying mutation and substitution processes, 
which are poorly understood but of independent interest due to their 
critical role in evolution. Studies of mammalian DNA by several 
investigators have revealed a number of substitution rate trends: 
transitions occur more frequently than transversions; in general G:C 
mutates to A:T at a higher rate than does A:T to G:C;  rates depend 
significantly on the flanking nucleotide context, with methylated C's 
in CpG dinucleotides being notable hotspots;  rates per year tend to 
be higher in lineages with short generation times, and higher in 
males than in females, suggesting that most mutations occur in 
conjunction with DNA replication (work of Li and colleagues); and 
recombination appears to play an important role in the neutral 
substitution process, likely reflecting the effects of biased gene 
conversion (recent work of Duret and colleagues).

Using orthologous mammalian sequences from NISC (www.nisc.nih.gov/), 
we have been working to better understand and model the 
characteristics of naturally occurring mutations, and to use this 
knowledge to more efficiently identify sequences under selection. We 
have found (Nat Genet 33, 514-517 (2003)) a mutational asymmetry 
associated with transcribed regions that likely reflects an 
interaction between DNA polymerase errors and transcription-coupled 
repair; this mutational asymmetry has acted over long evolutionary 
periods to produce a compositional asymmetry within most genes. More 
recently, Dick Hwang in my group has developed a powerful Bayesian 
Markov chain Monte Carlo approach to modeling sequence evolution that 
allows systematic exploration of context-dependent rates and 
mutational asymmetry by position within an evolutionary tree and 
within a sequence.  We have applied this to investigate variation in 
substitutional patterns in mammalian evolution, finding in particular 
that CpG mutations show a reduced generation time effect relative to 
other mutation types (Hwang and Green, PNAS 101, 13994-14001 (2004)). 
I will discuss recent extensions to this work allowing us to detect 
selected features and explore variation in selection strength in 
coding and non-coding sequences across the mammalian phylogeny. 

Host: Dr. Christopher B. Burge
Biology and Biological Engineering                          
Contact: Brenda Pepe, Biology, 452-3885
http://csbi.mit.edu/

Sponsored by  CSBi
Annual CSBi Seminar Series in Computational and Systems Biology

Directions can be found at the following url:
http://whereis.mit.edu/map-jpg
--