Trouble with edgeR (within Trinity wrapper)

[lysosome]

Hello everyone! I am much a neophyte to RNAseq analysis, so please bear with me.

I am trying to analyze an RNAseq dataset consisting of 3 treatments, with no replication. Each treatment has approximately 30 million reads.

The de novo assembly of all reads (combined from all treatments, following trimming) seemingly went smoothly. After mapping reads of each treatment to the transcriptome assembly using RSEM, I run into trouble.

I merged my 3 isoforms.results files using merge_RSEM_frag_counts_single_table.pl. After extracting transcript length information from one of the isoform.results files, I attempted to run edgeR, as follows:
------------
$ /home/user/software/trinityrnaseq_r2013_08_14/Analysis/DifferentialExpression/run_DE_analysis.pl --matrix /workarea/user/treatmenttrim/transcripts.counts.matrix --method edgeR --output /workarea/user/treatmenttrim/edgeR --no_eff_length
Unknown option: no_eff_length
Got 3 samples, and got: 4 data fields.
Header: treatment_notx.isoforms.results treatment_15.isoforms.results treatment_30.isoforms.results
Next: comp135535_c0_seq1 0.00 0.00 0.00

$VAR1 = {
'treatment_15' => 2,
'treatment_notx' => 1,
'treatment_30' => 3
};
$VAR1 = {
'treatment_15' => [
'treatment_15'
],
'treatment_notx' => [
'treatment_notx'
],
'treatment_30' => [
'treatment_30'
]
};
Samples to compare: $VAR1 = [
'treatment_15',
'treatment_notx',
'treatment_30'
];

CMD: R --vanilla -q < transcripts.counts.matrix.treatment_15_vs_treatment_notx.treatment_15.vs.treatment_notx.EdgeR.Rscript
sh: R: command not found
Error, cmd: R --vanilla -q < transcripts.counts.matrix.treatment_15_vs_treatment_notx.treatment_15.vs.treatment_notx.EdgeR.Rscript died with ret (32512) at /home/user/software/trinityrnaseq_r2013_08_14/Analysis/DifferentialExpression/run_DE_analysis.pl line 439.




WARNING: This EdgeR comparison failed...

CMD: R --vanilla -q < transcripts.counts.matrix.treatment_15_vs_treatment_30.treatment_15.vs.treatment_30.EdgeR.Rscript
sh: R: command not found
Error, cmd: R --vanilla -q < transcripts.counts.matrix.treatment_15_vs_treatment_30.treatment_15.vs.treatment_30.EdgeR.Rscript died with ret (32512) at /home/user/software/trinityrnaseq_r2013_08_14/Analysis/DifferentialExpression/run_DE_analysis.pl line 439.




WARNING: This EdgeR comparison failed...

CMD: R --vanilla -q < transcripts.counts.matrix.treatment_30_vs_treatment_notx.treatment_30.vs.treatment_notx.EdgeR.Rscript
sh: R: command not found
Error, cmd: R --vanilla -q < transcripts.counts.matrix.treatment_30_vs_treatment_notx.treatment_30.vs.treatment_notx.EdgeR.Rscript died with ret (32512) at /home/user/software/trinityrnaseq_r2013_08_14/Analysis/DifferentialExpression/run_DE_analysis.pl line 439.




WARNING: This EdgeR comparison failed...

------------

I imagine I have made a rather foolish error... Are the errors suggesting that R is not correctly installed? I would appreciate any assistance.

Thank you!

[Kennels]

yes, it cannot find your R installation. Did you install it?
If so, try putting it in your PATH.
e.g.

export PATH=/path/to/R:$PATH

where /path/to/R is where you installed R.

then type 'R' in your command line and see if it loads.
To exit R, type 'quit()'

[lysosome]

Hello Kennels,

Thanks for your help - R was installed, but foolishly/ignorantly was not in my PATH.