How to browse InterProScan, BLAST, HMMER results to retrieve relevant results?
I just started working on two already annotated transcriptomes; they have been annotated with BLASTP, InterProScan, HMMER and GO terms. I consistently find problems in retrieving functional information (e.g., retrieve all the endonucleases) from annotation results, and I always end up performing a keyword search in terminal with grep and similar tools, which gives me partial results.
Is there any smarter and more biologically correct way to browse annotation results?
• 3,105 views
•
link
updated
by
cdsouthan
•
written
by
keepclam •
•
written
by
keepclam •
1 answer
Log in to answer this question.
More posts like this
-
AGAT does not propagate DIAMOND functional annotations to BRAKER GFF3
written by Kinoppy •
-
Editing and Adding to a GTF file
written by Keem Uarren Eiryll •
-
Tutorial: Single-cell RNA-seq: Annotation: Celltype auto annotation with SCSA
written by Julia Ma •
-
NCBI Relevant files not being returned as query results. Trying to build exhaustive dataset.
written by chordata •
-
KEGG pathway annotation and enrichment analysis on an unannotated organism
written by Raito92
-
Finding overlaps between GRanges objects
written by mickey_95 •
-
Tutorial: Clustering of DAVID gene enrichment results from gene expression studies
written by Kevin Blighe
-
Tutorial: InsideDNA: RNA-Seq de novo assembly using Trinity
written by shilparaopradeep •
-
Job: Bioinformatics Programmer at Baylor College of Medicine, Houston TX
written by Vivek
-
Problem with different release of Ensembl
written by Lalla •
What format are the annotations in? Genbank/GFF or just text without format?
No, simple tab-separated text. One line of InterPro output looks like this:
while one line of HMMER output looks like this:
When you say you want to "browse" what are you expecting out of that? Do you need a summary of all different types of domains identified? Are you interested in knowing how many loci have no identifiable function?
Potentially you could use
awkto cut columns out of these file followed by some sort of sorting to classify the results.I'd like to retrieve all proteins belonging to a given group of interest. Let's suppose I want to retrieve all nucleases. If I <grep> "nuclease", I automatically exclude from my results all those nucleases that don't have "nuclease" in their annotation. Is there any means to circumvent this problem? Note that InterProScan and HMMER give database IDs of their results (Pfam for HMMER and varous dbs for InterProScan).