[BioC] gene2pathway retrain: which model is more complete?

Bogdan b.t.tokovenko at imbg.org.ua
Tue Aug 24 00:11:20 CEST 2010


Replying to the last one of my own questions:

> Of all the retrain()-generated models I now have, which one is
> theoretically better to use?
> The one with the most genes, most level detectors, or most features (domains)?
> Or the one with the lowest average prediction error, disregarding all
> other factors?

version    #genes        #domains        #detectors        avg.
error(ksvm) over 11 bags
-- not using custom gene2Domains --
1.6.1       3987/5577   3488                75                    0.006523582
1.6.0       4055/5667   3533                74                    0.007034632
-- using custom gene2Domains --
1.6.0       5667/5667   4007                78                    0.008454084

I've decided to use the last model.

-- 
Regards,
Bogdan



More information about the Bioconductor mailing list