Hello,
I am trying to generate a count matrix using abundance_estimates_to_matrix.pl script in Trinity, but I keep running into the same error. I fed the script two RSEM.genes.results files:
$ perl /usr/local/bin/trinityrnaseq-2.0.6/util/abundance_estimates_to_matrix.pl --est_method RSEM ~/aligning/aligning5/RSEM.genes.results ~/aligning/aligning6/RSEM.genes.results -reading file: /home/npetrill/aligning/aligning5/RSEM.genes.results -reading file: /home/npetrill/aligning/aligning6/RSEM.genes.results
- Outputting combined matrix.
Error, the column headings: RSEM.genes.results RSEM.genes.results are not unique. Should you consider using the --name_sample_by_basedir parameter? at /usr/local/bin/trinityrnaseq-2.0.6/util/abundance_estimates_to_matrix.pl line 174.
Can anybody make sense of this for me?
Thanks for your help, Nikelle
1 answer
I know it's probably too late to be helpful, but the individual RSEM.genes.results files need to each have a unique name if you're not using the --name_sample_by_basedir option.
In order for Trinity to use the R package edgeR, you have to install it. In R you can type:
source("http://bioconductor.org/biocLite.R")
Followed by
biocLite('edgeR')
Trinity was able to find edgeR in the default R library folder I use with no problem after I installed it this way.
Log in to answer this question.
Update: I added in --name_sample_by_basedir and got this error:
$ perl /usr/local/bin/trinityrnaseq-2.0.6/util/abundance_estimates_to_matrix.pl --est_method RSEM --name_sample_by_basedir ~/aligning/aligning5/RSEM.genes.results ~/aligning/aligning6/RSEM.genes.results -reading file: /home/npetrill/aligning/aligning5/RSEM.genes.results -reading file: /home/npetrill/aligning/aligning6/RSEM.genes.results
/usr/local/bin/trinityrnaseq-2.0.6/util/support_scripts/run_TMM_scale_matrix.pl --matrix matrix.not_cross_norm.fpkm.tmp > matrix.TMM.fpkm.matrixCMD: R --vanilla -q < __tmp_runTMM.R 1>&2