MirDeep 2 fail during execution of miRDeep2.pl

[RTK45]

Hi all,

i am using the script miRDeep2.pl, i use this command line :

miRDeep2.pl reads.fa genome.fa reads_vs_genome.arf none none none

At first, the program is running correctly, but when i came to the randfold p-values step i obtain this :

#computing randfold p-values
#computing randfold p-values
select_for_randfold.pl mirdeep_runs/run_27_03_2012_t_18_14_41/tmp/signature.arf mirdeep_runs/run_27_03_2012_t_18_14_41/tmp/precursors.str > mirdeep_runs/run_27_03_2012_t_18_14_41/tmp/precursors_for_randfold.ids

started: 18:15:23
Use of uninitialized value $id in hash element at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 617.
Use of uninitialized value $id in hash element at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 618.
Use of uninitialized value $id in hash element at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 619.
Use of uninitialized value $id in hash element at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 620.
Use of uninitialized value $db_old in hash element at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 260.
Use of uninitialized value $struct in length at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 261.
Use of uninitialized value $db_old in hash element at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 289.
Use of uninitialized value $db_old in hash element at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 290.
Use of uninitialized value $db_old in hash element at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 291.
Use of uninitialized value $seq in length at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 292.
Use of uninitialized value $query in hash element at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 424.
Use of uninitialized value $query in hash element at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 425.
Use of uninitialized value $query in hash element at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 437.
Use of uninitialized value $end in numeric le (<=) at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 682.
Use of uninitialized value $beg in numeric le (<=) at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 682.
Use of uninitialized value $seq in length at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 686.
Use of uninitialized value $beg in numeric le (<=) at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 686.
Use of uninitialized value $beg in subtraction (-) at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 689.
Use of uninitialized value $beg in subtraction (-) at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 689.
Use of uninitialized value $end in subtraction (-) at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 689.
Use of uninitialized value $seq in substr at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 689.
Use of uninitialized value $strand in string eq at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 692.
Use of uninitialized value $struct in length at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 707.
Use of uninitialized value $end in numeric le (<=) at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 710.
Use of uninitialized value $beg in numeric le (<=) at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 710.
Use of uninitialized value $struct in length at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 714.
Use of uninitialized value $beg in numeric le (<=) at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 714.
Use of uninitialized value $beg in subtraction (-) at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 717.
Use of uninitialized value $beg in subtraction (-) at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 717.
Use of uninitialized value $end in subtraction (-) at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 717.
Use of uninitialized value $struct in substr at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 717.
Use of uninitialized value $strand in string eq at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 450.
Use of uninitialized value $end in numeric le (<=) at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 682.
Use of uninitialized value $beg in numeric le (<=) at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 682.
Use of uninitialized value $seq in length at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 686.
Use of uninitialized value $beg in numeric le (<=) at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 686.
Use of uninitialized value $beg in subtraction (-) at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 689.
Use of uninitialized value $beg in subtraction (-) at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 689.
Use of uninitialized value $end in subtraction (-) at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 689.
Use of uninitialized value $seq in substr at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 689.
Use of uninitialized value $strand in string eq at /home/me/Desktop/mirdeep2/select_for_randfold.pl line 692.

Followed by other errors in the next scripts, i think my input files are ok.

Anyone have an idea ? i am lost

[gu_genetics]

I am having the same issues. Everything runs great until I get to this step.

#computing randfold p-values
select_for_randfold.pl mirdeep_runs/run_02_04_2012_t_14_10_05/tmp/signature.arf mirdeep_runs/run_02_04_2012_t_14_10_05/tmp/precursors.str > mirdeep_runs/run_02_04_2012_t_14_10_05/tmp/precursors_for_randfold.ids

started: 14:24:38
Use of uninitialized value in hash element at /data/lab/mirdeep2/select_for_randfold.pl line 260.
Use of uninitialized value in length at /data/lab/mirdeep2/select_for_randfold.pl line 261.

Any advice or direction would be greatly appreciated!

[aaronrjex]

I had this same problem. The issue seems to be a bug in the miRDeep2.pl with not providing any known miRNA sequences for the organism (or at least that was what was the problem for me). It seems to create a problem for the script that makes the signature.arf file (prepare_signature.pl)

So in the manual, the usage for this script is "prepare_signature.pl reads_collapsed.fa potential_precursors.fa -a miRBase_this_species.fa > signature.arf", where -a provides the sequence information for known microRNAs for the species in miRBase.

The problem is when you use the 'none' option in the command line (i.e., you have no known miRNAs for the species), the script should replace "-a XXX" with "none", but instead it just leaves this part of the command line blank, so you get "prepare_signature.pl reads_collapsed.fa potential_precursors.fa > signature.arf". This seems to be a problem even if you have provided miRNA sequences from a relate organism as an miRNA_other.fa file. If you cut and paste the command for this script from your log file and physically type in "none" after your precursors.fa file (i.e., "prepare_signature.pl reads_collapsed.fa potential_precursors.fa -a miRBase_this_species.fa none > signature.arf"), it should run the command (or at least it worked for me). Obviously that doesn't help for running the pipeline however. The only way I found to get around this was to give the program a dummy fasta file for the known microRNAs. In my case, I used a single predicted microRNA sequence (from the output files from the quantifier.pl module) with an identical match to a well known microRNA in model organism in miRBase. I imagine you could put any bit of sequence in there under the assumption that if the program can't find it, it won't matter. In either case, when you add this dummy file, the program seems to complete just fine. Someone with more programming nous than I (not a high bar that one) could probably work out a way to augment the miRDeep2.pl file to fix this error, but that's beyond my meagre powers.

Hope that helps.

[jay2008]

you may use miRDeep* where the genomic mapping is embedded.

http://nar.oxfordjournals.org/content/early/2012/12/04/nar.gks1187.abstract

[Latifa77]

Hi, Could anyone solve the problem of the script "select_for_randfold.pl"?? Because I have the same problem and I am working with the known miRNA sequences of my organism.