# No output in filter_P_prediction file

[baljinder-y]

I am trying to run bash script on test dataset using default command provided in the manual. I am getting no output in filter_P_prediction file. File is blank.
Moreover, the script_err file displays very low mapping percentage (0.05%) of sequencing data against mature miRNA sequences.
Kindly provide some solution

[alanlamsiu]

Hi @baljinder-y. Could you provide more details on which dataset and what commands did you use for testing?

[baljinder-y]

Hi @alanlamsiu,
Thank you for your time. I was trying to work with the test dataset i.e., GSM2094927.fa and TAIR10_genome.fa to test the pipeline.
First, I tried to run simple bowtie command to check alignment of reads against index of mature miRNAs provided with the tool.

This is the command that i used which is available in the bash script:
"bowtie -v 2 mature_index -f GSM2094927.fa > mir.aln"

This is the output:
reads processed: 968888
reads with at least one reported alignment: 1343 (0.14%)
reads that failed to align: 967545 (99.86%)
Reported 1343 alignments to 1 output stream(s)

Also i tried running bash script directly. This is the command that i was running.
./miRDP2-v1.1.4_pipeline.bash -g TestData/TAIR10_genome.fa -x TestData/TAIR10.genome -f -i TestData/GSM2094927.fa -o output/ -p 6

The command generates all the necessary files in the output folder. However, "GSM2094927_filter_P_prediction file" is blank.
I think issue is related to the low alignment percentage of reads with miRNA sequences.

[alanlamsiu]

Hi @baljinder-y,

The alignment rate of your read data to the reference genome is very low, which suggests either the reference genome does not match the read data, or there are some issues in the read data. It is necessary to inspect whether the read data is clean, e.g. free of adapter sequences. Or try to blast the reads against the reference genome and check whether they are the same species. I also notice that you use mature_index in your bowtie command. miRDP2 does require to align the read data to the genome sequence instead of mature miRNA sequence. You shouldn't proceed with miRDP2 before improving the alignment rate.