I have whole genome data in fasta format, and the gene/protein sequences of the target enzymes. I've tried BLAST for 2 sequences, but the results are strange, possibly because the whole genome sequence is too long as compared with the target gene? The Max Score is 37.4, Total Score is 1811, Query Cover is 45%, E value is 0.013, Identity is 92%, but basically what occurred was many short alignments along the genome, that do not seem to indicate the presence of the gene. How do people generally search for a gene in a genome, and how do you evaluate BLAST results? Isn't there a better method?
Above answer is good but as your blast image tells, you don't have any sequence similarity in both sequences. Thus, I don't think you will get any satisfactory results in any of your analysis. If you had any similarity you would have seen that in blast. So, its not the algorithm problem here, but its your sequence problem.
I would annotate first my genome which means predict all genes/proteins using an annotator. If you are not conformable with command line, you can use Galaxy.org. and the annotator selection depends on your organism.
Once you have your set of proteins, you can go for Blast Best hit (80 % identity and 70 % coverage) if you expect to have orthologuos or you can modify the Blast parameters otherwise.
In case you want only to blast against the genome, I would select BlastX and set the E-value to 10e-5 at least. The analysis should be careful as the sequences could share only part of the domain.
Hope it helps.
Best
- Badges
- Science topic
- Similar topics
- Mathematics
- Algebra
- Matrix