BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Using evolutionary sequence variation to build predictive models o
 f protein structure and function. - Lucy Colwell (University of Cambridge)
DTSTART:20161012T130000Z
DTEND:20161012T140000Z
UID:TALK68306@talks.cam.ac.uk
CONTACT:Emily Boyd
DESCRIPTION:The evolutionary trajectory of a protein through sequence spac
 e is constrained by its function. Collections of sequence homologs record 
 the outcomes of millions of evolutionary experiments in which the protein 
 evolves according to these constraints. The explosive growth in the number
  of available protein sequences raises the possibility of using the natura
 l variation present in homologous protein sequences to infer these constra
 ints and thus identify residues that control different protein phenotypes.
  Because in many cases phenotypic changes are controlled by more than one 
 amino acid\, the mutations that separate one phenotype from another may no
 t be independent\, requiring us to understand the correlation structure of
  the data.\n\nThe challenge is to distinguish true interactions from the n
 oisy and under-sampled set of observed correlations in a large multiple se
 quence alignment. We show that maximum entropy models of the protein seque
 nce\, constrained by the statistics of the multiple sequence alignment\, a
 re capable of predicting key aspects of protein function. These include (i
 ) the inference of residue pair interactions that are accurate enough to p
 redict all atom 3D structural models\; (ii) accurate predictions of bindin
 g partners between different proteins\; (iii) accurate prediction of bindi
 ng between protein receptors and their target ligands. We will discuss how
  a mathematical framework based on random matrix theory bounds which seque
 nce alignments contain sufficient information to build accurate predictive
  models. Finally\, we will pose questions about the physics of binding int
 eractions in an example from the immune system where large sets of evoluti
 onarily related sequences are not available. 
LOCATION:MR4\, Centre for Mathematical Sciences\, Wilberforce Road\, Cambr
 idge
END:VEVENT
END:VCALENDAR