Systematic quantitative function annotation of mouse genes

Several years after sequencing human and mouse genomes, much remains to be discovered about the functions of most genes. Computational prediction of gene function promises to help focus limited experimental resources on the most likely hypotheses. In this study, a standardized collection of mouse functional genomic data was assembled, and nine bioinformatics teams used this dataset to independently train classifiers and generate predictions of function for 21,603 mouse genes. We compared the performance of function prediction algorithms and identified strengths and weaknesses of current functional genomic datasets. Quantitative function predictions may be useful, e.g., in the study of complex disease in human populations.