Hello,
I'd like to know your opinions on identifying fungal DNA sequences using the BLASTn with regards to Max Score, Total Score, Query Cover, E-value, and Identity match %? What are common shortcomings of these categories?
One mycorrhizal study I'm reading selected their results from percentages Query-Cover, and Identity that equaled or surpassed 80%. Do these values follow the literature's general consensus (if there is one) on the matter?
Would thresholds differ (i.e., broaden or narrow) if the sequence best matched cryptic and/or uncultured taxa?
Thank you,
Gary
Dear Gary
unfortunately I don't think there is a general consensus about the thresholds to apply to BLAST or other similarity searches. A filter on query coverage percentage and identity of at least 80% sounds reasonable as well as focusing on the hits with an evalue less than 0.05. In the end it depends on the kind of information you are looking for, the distance between your species and the others in the database, etc. So in the end you decide if you want to be very strict applying stricter filters because you want to find the strongest candidate or if you want to apply more relaxed filters. For instance, if you wanted to collect similar sequences in order to perform a phylogenetic analysis or a multi-alignment then you could decide to be more relaxed to get most of the sequences. On the other hand if you are looking for a candidate gene for a specific trait then you could be more strict.
I hope this information might be useful.
Kind regards
Riccardo
Bioinformatics data analyst at www.sequentiabiotech.com
- Badges
- Science method