search

HenrikBengtsson / aroma.affymetrix

🔬 R package: Analysis of Large Affymetrix Microarray Data Sets
https://cran.r-project.org/package=aroma.affymetrix
10 stars 3 forks source link

TESTS: "addons" tests for GSE12702 runs 40 samples and takes ~3 hours - can we use fewer to speed it up? #14

Closed HenrikBengtsson closed 8 years ago

HenrikBengtsson commented 8 years ago

In:

  1. testScripts/addons/chipTypes/Mapping250K_Nsp/
    • 21.doASCRMAv2,CalMaTe.R
    • 31.doASCRMAv2,PSCBS.R
  2. testScripts/addons/chipTypes/Mapping250K_Nsp,Sty/
    • 11.doASCRMAv2,TCNs,500K.R

we use 20 tumor-normal pairs from the GEO dataset GSE12702.

The first set of tests takes nearly 3 hours to complete (sequentially). The AS-CRMAv2 step on the 40 arrays take a good 2 hours. The CalMaTe processing takes ~40 minutes for the 40 arrays. (The PSCBS is done on a single tumor-normal pair and a single chromosome and takes on a couple of minutes). Can we cut down this data set to, say, 10 tumor-normal pairs to speedup up these redundancy tests?

BTW, it turns out that I've disable the 2nd set of tests for some reason; maybe because there were race conditions where the two of them tried to run the same AS-CRMAv2 pipeline on the same set of arrays at the same time?

HenrikBengtsson commented 8 years ago

Updated to use only 12 arrays. The AS-CRMAv2 step now takes ~35 mins (was ~2 hours). The CalMaTe processing takes ~30 minutes (was ~40 minutes) for the 12 arrays.