What Would Be The Next Analysis Steps After Producing Alignments With Blastx
Hello, I have 200000 seq-s. I run blastx against a protein database, filtered the hits by identity threshold >90%. Can anybody suggest what to do next?I need to annotate those seq-s. Any ideas how to further analyze blast output results?
• 547 views
•
link
updated
by
Istvan Albert
•
written
by
User000
•
written
by
User000
0 answers
No answers yet.
Log in to answer this question.
More posts like this
-
News: ClusteredNR NCBI BLAST database provides faster searches and more informative BLAST results
written by PeterC_NCBI •
-
Annotate coding sequences
written by karthic
-
How to add annotation to a list of Genbank IDs (obtained from BLAST)?
written by xupbuy •
-
Output blastx top hits as fasta of the original sequences
written by bonsalldavid •
-
Problems with accessing the Swissprot database using qblast
written by bondserv •
-
Speeding Up The Blast Job
written by User000
-
Plant Annotation Workflow
written by User000
-
Filtering Spliced Alignment Before Build Gene Models
written by Plantae
-
Orf Finder Vs Blastx
written by User 3822 •
-
Merging Blastx Hits From Overlapping Bacterial Genome Segments
written by Darked89
You can simply take n-top hits for each of your sequences and to synthesize an annotation that combines n-annotations. Or, alternatively, you can load the whole blastx output into MEGAN in order to partition your set by taxonomical assignment or by function, export subsets and assign them annotation too, in this case, you will rely on megan's "voting" and will a have nice visualization.
Never heard of megan, but will try. Thank you for a useful tip!
As you said below you have contigs you have to be careful with Megans visualization. A species with five 500bp contigs assigned to it will look the same as a species with five 10,000bp contigs assignen to it. Or at least it did, when I last used it.
The further analysis completely depends on what information you wanted to get out of what kind of sequence data. Is this data from one genome, an environmental sample, are those raw reads, assembled sequences, gene sequences?
they are assembled contigs. I heed to also analyse isoforms and homeologs, which I suppose I can see from blast output? in blast I sorted proteins in this way that I can see all contigs that align to it, always with >90 identity threshold, I guess I can put those contigs together then? sorry for stupid q-s, just never worked with this things
I'd suggest that you think long and hard what it is that you want to find out.
I need to annotate them,my final goal,as I have mentioned in my question, but not sure which step to take further,If I knew I wouldn't be posting a question here. Suggestions are welcomed!
Annotate the contigs with best hits (vs your db) to each segment? In what file format? You want to write this info to the fasta headers or what? If you blasted against refseq, you could easily get some related annotations like KOs or ECs or whatever. Which genetic code did you use with blastx? I assume this setting affects your results to some degree..
I had assembled transcripts, I run them in blastx against protein database (i.e.swissprot) and retrieved only hits with >90% id,I need to functionally annotate them knowing that my assembled contigs come from a plant