Functional class predictions for ORFs in the yeast genome

These are the predictions made for S. cerevisiae ORFs belonging to classes "99: UNCLASSIFIED PROTEINS" and "98: CLASSIFICATION NOT YET CLEAR-CUT". The classes were taken from the MIPS functional catalog on 24/4/02, and are listed in the file classes.txt. The predictions are described in the PhD thesis "Machine learning and data mining for yeast functional genomics", Amanda Clare, UWA (February 2003, 1Mb pdf) and in Clare, A. and King R.D. (2003) Predicting gene function in Saccharomyces cerevisiae. 2nd European Conference on Computational Biology (ECCB '03). (published as a journal supplement in Bioinformatics 19: ii42-ii49). See also the results of learning with phenotype data and results from experiments clustering Guilt By Association data. And all data sets used in these experiments can be found here.

For a more user-friendly representation of these predictions go to http://www.genepredictions.org.

The predictions can be downloaded as individual files, or as a single bzipped tar file containing everything (1.4M).

dataset	Level 1	Level 2	Level 3	Level 4	Level all (0)
seq	1	2	3	4	0
pheno	1	2	3	4	0
struc	1	2	3	4	0
hom	1	2	3	4	0
expr	1	2	3	4	0
cellcycle	1	2	3	4	0
church	1	2	3	4	0
derisi	1	2	3	4	0
eisen	1	2	3	4	0
gasch1	1	2	3	4	0
gasch2	1	2	3	4	0
spo	1	2	3	4	0
ceho	1	2	3	4	0
ceph	1	2	3	4	0
cese	1	2	3	4	0
cest	1	2	3	4	0
seho	1	2	3	4	0
seph	1	2	3	4	0
sest	1	2	3	4	0
stho	1	2	3	4	0
phho	1	2	3	4	0
phst	1	2	3	4	0
all	1	2	3	4	0