Format

Send to

Choose Destination
Gene. 2003 Oct 30;318:169-75.

Human disease genes: patterns and predictions.

Author information

1
Department of Evolutionary Biology, Evolutionary Biology Centre, Uppsala University, Norbyvägen 18D, 752 36 Uppsala, Sweden. nick.smith@ebc.uu.se

Abstract

We compared genes at which mutations are known to cause human disease (disease genes) with other human genes (nondisease genes) using a large set of human-rodent alignments to infer evolutionary patterns. Such comparisons may be of use both in predicting disease genes and in understanding the general evolution of human genes. Four features were found to differ significantly between disease and nondisease genes, with disease genes (i) evolving with higher nonsynonymous/synonymous substitution rate ratios (Ka/Ks), (ii) evolving at higher synonymous substitution rates, (iii) with longer protein-coding sequences, and (iv) expressed in a narrower range of tissues. Discriminant analysis showed that these differences may help to predict human disease genes. We also investigated other factors affecting the mode of evolution in the disease genes: Ka/Ks is significantly affected by protein function, mode of inheritance, and the reduction of life expectancy caused by disease.

PMID:
14585509
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center