EXPANDS 2.1 release
72 views
Skip to first unread message
Noemi Andor
Feb 24, 2018, 6:33:17 PM2/24/18
to EXPANDS
EXPANDS 2.1.0 is now available on CRAN:
https://cran.r-project.org/web/packages/expands
-----------------------------------------
CHANGES IN EXPANDS VERSION 2.1.0
NEW FEATURES
o Included parameter ploidy in runExPANdS.R, allowing specification of non-diploid background states (e.g. for near-triploid cell lines).
o Parallel computing option for runExPANdS.R (parameter nc).
BUG FIXES
o Fixed undesired behaviour in Subpopulations.java where integer division was rounded thereby rejecting valid mutated and wild type copy number configurations.
o Fixed bug in runExPANdS.R where the number of mutations per subpopulation was recorded incorrectly in .spstats output.
o Fixed axis-range bug in plotSPs.R.
OTHER CHANGES
o Copy-neutral LOH are no longer modelled.
o Fit of copy number status has a higher weight (x^2) than fit of point mutation status (x^1).
o Likelihood of mutation events is no longer higher in cancer-cell populations known to be non-founders.
o Binary option 'ambig' in assignQuantityToSP.R replaced with parameter 'e', allowing user to finetune amount of accepted ambiguity within a continuous range (e==0 is equivalent to former ambig==F, while e>0 corresponds to ambig==T).
https://cran.r-project.org/web/packages/expands
-----------------------------------------
CHANGES IN EXPANDS VERSION 2.1.0
NEW FEATURES
o Included parameter ploidy in runExPANdS.R, allowing specification of non-diploid background states (e.g. for near-triploid cell lines).
o Parallel computing option for runExPANdS.R (parameter nc).
BUG FIXES
o Fixed undesired behaviour in Subpopulations.java where integer division was rounded thereby rejecting valid mutated and wild type copy number configurations.
o Fixed bug in runExPANdS.R where the number of mutations per subpopulation was recorded incorrectly in .spstats output.
o Fixed axis-range bug in plotSPs.R.
OTHER CHANGES
o Copy-neutral LOH are no longer modelled.
o Fit of copy number status has a higher weight (x^2) than fit of point mutation status (x^1).
o Likelihood of mutation events is no longer higher in cancer-cell populations known to be non-founders.
o Binary option 'ambig' in assignQuantityToSP.R replaced with parameter 'e', allowing user to finetune amount of accepted ambiguity within a continuous range (e==0 is equivalent to former ambig==F, while e>0 corresponds to ambig==T).
Reply all
Reply to author
Forward
0 new messages