Discripancy in miRDeep2.pl and quantifier.pl outputs
1
1
Entering edit mode
11 months ago
otieno43 ▴ 40

I am working with miRNA-seq dataset from a non model organism with no known miRNA. As such I have identify both known and novel miRNAs and I am using miRDeep2. I have five replicates for each condition. Because my organism has no annotated miRNA, I used mature miRNA from a close relative (D. melanogaster) to my organism to identify known and novel miRNAs. To do this, I combined all the sequencing data into one and used the code below for identification of miRNAs.

miRDeep2.pl A_collapsed.fa A.fasta A_collapsed_vs_genome.arf none miRBase_dme_v14.fa none 2>A_report.log

This identified known and novel miRNAs with their associated Total and unique number of reads that mapped to them. Sorted the miRNA based on the miRDeep2 score >2, randfold p-value = Yes, and at least 10 reads mapping. This now acts as my reference.

How when I used the mature miRNA, Star consensus, and precursor sequences from the miRDeep2 output (my reference) to run quantifier.pl for miRNA expression for each replicates, I run into a problem where some of the miRNAs with thousands of reads (over 1000 counts) mapping to them have zero reads (zero counts) with quantifier.pl. The code for quantifier.pl:

quantifier.pl -p ./for_qiantifier/precursor.fasta -m ./for_qiantifier/Mature.fasta \
-r ./GffI_cat-reads_collapsed.fa -s ./for_qiantifier/Star.fasta -y now

I am wondering if this is normal or there is something wrong somewhere. Some insight please.

Again, when I run initial code for miRDeep2.pl for each replicate, I realize that some mature miRNA are identified in different parts of the genome different from the one identified when combined seq data is used. How do I reconcile this.

Finally in my output results for both miRDeep2 and quantifier.pl, the .cvs files do not have input IDs or species IDs if I just put some. I am wondering if this should be the case now that the organisms I am working with has no known miRNAs.

Finally, final, I also get something that looks like ‘error’ but not when I run quantifier.pl:

Can't use string ("45") as a HASH ref while "strict refs" in use at /home/ea432/.conda/envs/umi_mir/bin/make_html2.pl line 658.

Can't use string ("57") as a HASH ref while "strict refs" in use at /home/ea432/.conda/envs/umi_mir/bin/make_html2.pl line 658. Etc.

Some help please. Thanks.

miRDeep2 • 1.1k views
ADD COMMENT
0
Entering edit mode
9 months ago

Hello, have you solved it? I also encountered a similar problem. I found that some reads could not match the 3' end of the precursor, so the 3p of many mirnas could not be identified

ADD COMMENT
0
Entering edit mode

Sorry bingyanzheng18 I was not able to solve it. but when I eliminated -s Star.fasta it improved the mapping counts. You can email the authors of mirdeep2 for help. They should be able to help.

ADD REPLY

Login before adding your answer.

Traffic: 5064 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6