The Result Of The Mirdeep2

Question

[hua.peng1314]

Hi all.I am dealing with some miRNA data sequenced by illumina Htseq 2000.And I confuse the result of the mirdeep2 software. First is the result of the quantifier.pl.In the miRNAs_expressed_all_samples_time.csv has entries show below.

hsa-miR-1       549.00  hsa-mir-1-1     549.00  420.00  129.00  19.07   23.63
hsa-miR-1       564.00  hsa-mir-1-2     564.00  432.00  132.00  19.62   24.18

The same miRNA with different precursor.I am not sure how to deal with these when I try to use the R package DEGseq to detect DE miRNA.Treat them as two different entries or just plus them if these entries have no intersection.

Second,the result of less output.mrd |grep '>'|wc -l is 1789.I think it hsa 1789 precursors,and the result.csv file just have 785 novel and known miRNAs, I don't know how to explain the different between these two numbers. Your reply will be appreciated. Cheers.

[selkash2016]

What is the best pipeline for miRNA Differential expression analyssis, miRNA predicted from mirdeep2 software

Answer 1

[IV]

Different miRNA precursors can produce the same mature form.

In such cases, it depends on the type of your downstream analysis.

If you are interested in precursor expression then these two (and all others that fall in this category) should be considered as separate features.

If your analysis focuses on mature miRNAs (for target identification, etc) then these two matures should be considered as one, since they have the same mature sequence (same targets, etc).

Depending on your miRDeep 2 settings (if you divide multimaps or not), you can just add the expression of those two mature miRNAs into one, or keep one with just their average.

Cheers,

IV

[hua.peng1314]

Really help.Thank you very much. Cheers

[hua.peng1314]

There is no option for me to set to divide multimaps or not,and the document says "The module maps the deep sequencing reads to predefined miRNA precursors and determines by that the expression of the corresponding miRNAs. First, the predefined mature miRNA sequences are mapped to the predefined precursors. Optionally, predefined star sequences can be mapped to the precursors too. By that the mature and star sequence in the precursors are determined. Second, the deep sequencing reads are mapped to the precursors. The number of reads falling into an interval 2nt upstream and 5nt downstream of the mature/star sequence is determined." So I think it is multimaps as default and I should choose to keep one with just their average

[IV]

Following your read mapping with mapper, you can set division of multimaps in the quantifier.pl with the -W option.

I find it really useful.

[hua.peng1314]

You are right,I was careless and just read the document but not the "quantifier.pl -h".So I miss the option.Thank you very much

Answer 2

[David Langenberger]

Concerning your first question: Without being familiar with the miRDeep2 output format, hsa-miR-1 seems to be the family and hsa-mir-1-1 and hsa-mir1-2 two microRNA genes. Actually, if it says -1 and -2 these should be two copies of mir-1. If that is correct (you can check that by reading the miRDeep documentation and by checking miRBase), you can treat them as two different microRNAs in your downstream analysis.

[hua.peng1314]

Thanks for friendly to reply .But actually hsa-miR-1 is the microRNA gene and the hsa-mir-1-1 and hsa-mir1-2 are two precursors.I have see the papers and the documents of mirdeep and mirdeep2.What confuses me is I am not sure if the two numbers(549,564) have overlap.I think I should read the script though it's really a headache for me,Thanks again. cheers

[David Langenberger]

You are right, with naming hsa-miR-1 the family, I was wrong. has-miR-1 is a mature sequence. And the mature sequence hsa-miR-1 appears in two different stem-loops (precursors), namely hsa-mir-1-1 and hsa-mir1-2. The family is called mir-1 and consists of has-mir-1-1, has-mir-1-2 and has-mir-206 (in human). So, when looking at the expression of has-miR-206 in the stem-loop hsa-mir-206, you should find a similar value as for hsa-mir-1-1 and has-mir-1-2.