blast result against rfam database

[lran2008]

Hi All,

I am blasting my miRNA-seq fasta file against rFam data base. Here is the command line I used:
blastn -query TOTAL_mapped_reads.fa -task megablast -db Rfam.fasta -out mapped_vs_rfam.txt -outfmt "6 qseqid sseqid qlen evalue pident nident mismatch qcovs qcovhsp"

Then I want to retrive those reads which match tRNAs in rFam.
There are many hits in the output (see the attached picture).

But when I pick the first read which is shown to be 100% match with rRNA in rfam by searching it in rfam website, there is no hit.
>62-796337
TCCCATATGGTCTAGCGGTTAGGATTCCTGGTT

Briefly, according to my blast result against rfam database, there is a match, but when I search the my sequence in online rfam website, there is no hit. So what is the problem?

[rhinoceros]

Based on blastn (at ncbi) your sequence is not a tRNA. The 'best' hit is "PREDICTED: Dasypus novemcinctus WW domain-binding protein 4-like (LOC101430251), transcript variant 2, mRNA".

Perhaps consider a more conservative evalue cutoff? Also, how did you create your Rfam db? "-db Rfam.fasta" looks rather wrong to me but maybe you just named it oddly...

[lran2008]

Originally Posted by rhinoceros

Based on blastn (at ncbi) your sequence is not a tRNA. The 'best' hit is "PREDICTED: Dasypus novemcinctus WW domain-binding protein 4-like (LOC101430251), transcript variant 2, mRNA".

Perhaps consider a more conservative evalue cutoff? Also, how did you create your Rfam db? "-db Rfam.fasta" looks rather wrong to me but maybe you just named it oddly...

I indeed named it oddly. Anyway, it produced results, but seems to be unreliable... The e value for 62-796337 is 1e-10 in the attached picture. So what's your suggestion for the e value cutoff?