search

mskcc / facets

Algorithm to implement Fraction and Copy number Estimate from Tumor/normal Sequencing.
144 stars 67 forks source link

Ploidy not updated while tracing emcncf #129

Closed dvenet closed 5 years ago

dvenet commented 5 years ago

The ploidy estimate given is not updated during the tracing - for instance:

fit=emcncf(oo, min.nhet=tty[1], maxiter=10, trace=TRUE); iter: 1 dif: 0.1709942 purity: 0.7877296 ploidy: 2 iter: 2 dif: 0.07670063 purity: 0.7877296 ploidy: 2 iter: 3 dif: 0.01715529 purity: 0.7877296 ploidy: 2 iter: 4 dif: 0.01539139 purity: 0.7877296 ploidy: 2 iter: 5 dif: 0.007043781 purity: 0.7877296 ploidy: 2 iter: 6 dif: 0.005353876 purity: 0.7877296 ploidy: 2 iter: 7 dif: 0.005224373 purity: 0.7877296 ploidy: 2 iter: 8 dif: 0.004832115 purity: 0.7877296 ploidy: 2 iter: 9 dif: 0.002994015 purity: 0.7877296 ploidy: 2 iter: 10 dif: 0.002740225 purity: 0.7877296 ploidy: 2 fit$purity [1] 0.7877296 fit$ploidy [1] 4.196863

shenmskcc commented 5 years ago

Ploidy is not updated in the EM algorithm for the integer copy number calls. It is calculated at the end based on the 2-copy state (dipLogR) and the purity estimate from the EM update.

Ronglai

On Mon, May 6, 2019 at 2:30 PM dvenet notifications@github.com wrote:

The ploidy estimate given is not updated during the tracing - for instance:

fit=emcncf(oo, min.nhet=tty[1], maxiter=10, trace=TRUE); iter: 1 dif: 0.1709942 purity: 0.7877296 ploidy: 2 iter: 2 dif: 0.07670063 purity: 0.7877296 ploidy: 2 iter: 3 dif: 0.01715529 purity: 0.7877296 ploidy: 2 iter: 4 dif: 0.01539139 purity: 0.7877296 ploidy: 2 iter: 5 dif: 0.007043781 purity: 0.7877296 ploidy: 2 iter: 6 dif: 0.005353876 purity: 0.7877296 ploidy: 2 iter: 7 dif: 0.005224373 purity: 0.7877296 ploidy: 2 iter: 8 dif: 0.004832115 purity: 0.7877296 ploidy: 2 iter: 9 dif: 0.002994015 purity: 0.7877296 ploidy: 2 iter: 10 dif: 0.002740225 purity: 0.7877296 ploidy: 2 fit$purity [1] 0.7877296 fit$ploidy [1] 4.196863

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/mskcc/facets/issues/129, or mute the thread https://github.com/notifications/unsubscribe-auth/AC7CKVAOGNOU3W7ECBEUBVTPUB2ODANCNFSM4HLCKXNA .