Properties of biological fitness landscapes are of interest to a wide sector of the life sciences, from ecology to genetics to synthetic biology. For biomolecular fitness landscapes, the information we currently possess comes primarily from two sources: sparse samples obtained from directed evolution experiments; and more fine-grained but less authentic information from 'in silico' models (such as NK-landscapes). Here we present the entire protein-binding profile of all variants of a nucleic acid oligomer 10 bases in length, which we have obtained experimentally by a series of highly parallel on-chip assays. The resulting complete landscape of sequence-binding pairs, comprising more than one million binding measurements in duplicate, has been analysed statistically using a number of metrics commonly applied to synthetic landscapes. These metrics show that the landscape is rugged, with many local optima, and that this arises from a combination of experimental variation and the natural structural properties of the oligonucleotides.

Original publication

DOI

10.1098/rsif.2009.0193

Type

Journal article

Journal

Journal of the Royal Society, Interface

Publication Date

03/2010

Volume

7

Pages

397 - 408

Addresses

Manchester Interdisciplinary Biocentre, University of Manchester , 131 Princess Street, Manchester M1 7DN, UK. william.rowe@manchester.ac.uk

Keywords

Phycocyanin, DNA-Binding Proteins, Computational Biology, Adaptation, Biological, Evolution, Molecular, Models, Genetic, Aptamers, Nucleotide