Hello Can you tell me how many covariates you have in mod/mod0? And can you give me an indication of the sample size. The most common cause of this problem is that there are too many covariates in the model (mod) or there are co-linear (i.e. highly correlated) variables in mod. Best Jeff On Sat, Apr 29, 2017 at 6:19 AM amoltej [bioc] <noreply@bioconductor.org> wrote: > Activity on a post you are following on support.bioconductor.org > > User amoltej <https: support.bioconductor.org="" u="" 7192=""/> wrote Question: > SVA behaving weirdly, system is computationally singular , problem with > method leek only <https: support.bioconductor.org="" p="" 95383=""/>: > > I am running SVA on my data to remove the batch effect from the data that > I got from NCBI and data I sequenced. > > I am running following command - > > > svobj = sva(raw.data,mod,mod0,n.sv=n.sv) > Number of significant surrogate variables is: 3 > Iteration (out of 5 ):1 > Error in solve.default(t(mod) %*% mod) : > system is computationally singular: reciprocal condition number = 5.55112e-18 > Error during wrapup: cannot open the connection > > This problem is happening only when I am running "leek" method. When I am > using "be" method everything works fine. can somebody help me to solve this > mystery? > > Thank you > ------------------------------ > > Post tags: SVA, combat sva, leek > > You may reply via email or visit SVA behaving weirdly, system is computationally singular , problem with method leek only >