Dear teachers, I am studying some mirnas in the exosomes of a fish. I want to predict the target genes of these mirnas by using three kinds of software (miranda, targetscan, and rnahybrid). The data I input is the fasta file of the mirna sequence and the 3'utr sequence of each gene. But the results predicted by these three software are very different, can you help me analyze it? I can show my code.

miranda MvsB.fa 3UTR.fa -sc 150 -en -20 | grep '^>>'|sed 's/^M//g'| sed 's/>>//g' >miRanda_MvsB.txt perl targetscan_70.pl MvsB.txt 3UTR..txt MvsB_targetscan.txt RNAhybrid -c -s 3utr_human -t MvsB.fa -q 3UTR.fa |awk -F ":" '{print $1"\t"$3}'| uniq > MvsB_RNAhybrid_results.txt