Error in edgeR

[Yogesh Gupta]

Dear All,

Greetings!

when I am running this cammand getting some error

/home/yog/software/TRINITY_HOME/Analysis/DifferentialExpression/run_DE_analysis.pl --matrix /home/yog/matrix.counts.matrix --method edgeR

Trinity error

Got 4 samples, and got: 5 data fields.

Header: sp_log  sp_hs   sp_ds   sp_plat

Next: TRINITY_DN7231_c0_g1_i1   1.00    0.00    1.00    0.00

$VAR1 = {

          'sp_hs' => 2,

          'sp_plat' => 4,

          'sp_log' => 1,

          'sp_ds' => 3

        };

$VAR1 = {

          'sp_hs' => [

                       'sp_hs'

                     ],

          'sp_plat' => [

                         'sp_plat'

                       ],

          'sp_log' => [

                        'sp_log'

                      ],

          'sp_ds' => [

                       'sp_ds'

                     ]

        };

Contrasts to perform are: $VAR1 = [

          [

            'sp_ds',

            'sp_hs'

          ],

          [

            'sp_ds',

            'sp_log'

          ],

          [

            'sp_ds',

            'sp_plat'

          ],

          [

            'sp_hs',

            'sp_log'

          ],

          [

            'sp_hs',

            'sp_plat'

          ],

          [

            'sp_log',

            'sp_plat'

          ]

        ];

Error, cannot calculate dispersions due to lack of replicates. Specify a dispersion parameter --dispersion <float>. See help for details

I also want to know how we can do differential expression analysis when we have more than one biological replicate.

Thanks

--

Yogesh Gupta

Senior Research Fellow

National Agri-Food Biotechnology Institute
(Departement of Biotechnology, Government of India)

C-127, Industrial Area, Phase 8, SAS Nagar

Mohali-160071 Punjab, India

Cell No. +91-9041074172

[Brian Haas]

Without biological replicates, you need to set the dispersion parameter:  --dispersion

See the edgeR manual for info on this. If you want to just get it to work w/o error, try setting

   --dispersion 0.1

but for proper study, explore the manual.

best,

~brian

 

--
You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-users+unsub...@googlegroups.com.
To post to this group, send email to trinityrnaseq-users@googlegroups.com.
Visit this group at https://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

--

--
Brian J. Haas
The Broad Institute
http://broadinstitute.org/~bhaas