Regensburg 2016 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

BP: Fachverband Biologische Physik

BP 27: Computational Biophysics

BP 27.5: Talk

Tuesday, March 8, 2016, 11:00–11:15, H43

Inferring Co-evolution in proteins and RNA by Maximum Entropy Based Approaches — •Alexander Schug — Karlsruher Institut für Technologie, Steinbuch Centre for Computing

Protein function often requires a protein to form a complex or adopt multiple conformations during its functional cycle. The increasingly ubiquitous availability of sequential information for many protein families has given rise to a Maximum Entropy based approach called Direct Coupling analysis [1], which traces amino acid co-evolution to extract contact maps out of only sequence information. This is sufficient information for the blind prediction of quaternary and tertiary protein [2,3] or RNA structures [4]. Residue co-evolution therefore guarantees the structural stability of a protein including its functional conformations. Similarly, we can infer mutational landscapes and capture epistatic couplings between residues, and assess the dependence of mutational effects on their sequence context [5]. We find an about 40% in explicative power as compared to approaches neglecting epistasis.

References

[1] Weigt M et al., PNAS (2009) 106, 67-72; F. Morcos et al., PNAS (2011) 108, E1293-E1301

[2] Schug A et al., PNAS (2009) 106, 22124-22129

[3] Dago A et al., PNAS (2012), 109: E1733-42

[4] De Leonardis E et al., NAR (2015), doi: 10.1093/nar/gkv932

[5] Figliuzzi M et al., Mol Biol Evol (2015), doi: 10.1093/molbev/msv211