Stucture and Function

D. Structure-function relationships:
Sequence patterns that predict function

One of the most challenging areas of computational molecular biology is the prediction of the function of protein molecules from their sequence.

sequence determines 3-D structure, structure determines function

Cannot predict a 3-D protein structure from amino acid sequence alone. Best current approach relies on sequence similarity to proteins of known structure = "threading"

Can predict some aspects of 3-D structure from sequence:

A-helix vs. B-sheet
membrane spanning region
helix-turn-helix
signal peptide

identifying conserved regions (domains or motifs)

functions of these conserved domains are defined by laboratory research

Domain databases can be used to scan any unknown protein sequence for the presence of over a thousand known domains

Similarly, databases of important conserved elements within DNA sequences have been developed:

transcription factor binding sites
restriction enzyme recognition sites

Some 3D RNA structures can also be predicted based strictly on sequence

by sequence comparison with other known sequences (such as tRNA)
or by simple detection of stem-loop structures as inverse repeats

Using Computers for Molecular Biology
Stuart M. Brown, Ph.D., RCR, NYU Medical Center
Comments to: browns02@mcrcr.med.nyu.edu

D. Structure-function relationships: Sequence patterns that predict function

D. Structure-function relationships:
Sequence patterns that predict function