I have transcriptomic data for four plant species under certain growth condition. I need to find the genes that are common to all four species. I have generated the transcript sequences from the raw RNASeq data. I performed an all-vs-all BLAST search by the BLAST+ package. Next I need to parse the BLAST output to perform reciprocal/bidirectional BLAST to prepare the orthologous table for the four plants. How and what sets of commands do I need to use to perform the task? Further, how can I pull out the best sequences from the blast output (in STEP 1 above)? My input was a formatted BLAST database of all the transcripts of all the four plant species, inputs were multifasta formatted files of the transcripts for each of the plant for each search.
Hi Javed,
I would use orthomcl for this task. It takes your blast and will return you groups of genes with the best reciprocal blast... I personally prefer version 1.4, as it can be run very quickly.
Best,
Thomas
I would also use OrthoMCL or OMA (https://omabrowser.org/standalone/).
Thank you Dr. Thomas and Dr. Nicola for your valuable suggestions.
orthoMCL considers only protein sequences as inputs. But what I have are nucleotide sequences. I am also not sure if they are full open reading frames, since they have been assembled from RNA-seq data. Therefore, before using orthoMCL, I need to assign each nucleotide sequence (in multi fasta files) of each species (total 4 plants species) to a true protein sequence (not just predicted ORF).
Please suggest. Thanks.
Ajit kumar Roy
Thank you sir for your kind appreciation, Instead of explanation I need "command flow" for all vs all bastPlease help
Javed ChoudharyDid you use custom database to run your nt query sequences against or did you use the formatted nt databses from the ncbi ftp site?
- Badges
- Science topic
- Similar topics
- Engineering
- Modeling