As stated, I would like to know what is the difference between NCBI blast and EzBioCloud?
I have sequences of multiple unknown bacteria and am trying to do the first identification on them using only the 16s sequence by sanger sequencing. Once they have been screened, I will pick the one that looks interested to move forward with the characterization.
However, when I tried to use these two services, I have gotten a bit different results.
For example, in NCBI blast, the sample I have generated more than 10 hits that have < 99% identity. However, in EzBioCloud, I can only get one hit up to 98.99% of similarity and it doesn't have a strain name or taxon name. The highest one with strain name and taxon name only has 97.91% similarity. Therefore, I don't know if this one can be considered a potential novel strain or if someone might already have isolated and identity it.
I was wondering what's the difference? Thank you.
What is the difference
- difference sequence data set as database
- different search algorithm, parmeters
- taxonomic variations
what you are dealing with is a very common issue, please read more about it.
One of the problems with sequencing is that there are many databases with different levels of completeness and coverage. My suggestion is to stick with NCBI. It is the primary source for sequence identification.
As I am sure you are aware that no two genetic sequences are 100% identical. Unless your 1% of differences has profound phenotypical outcomes... I would trust the data you are presented with.
1. If you did not use "type materials only" option, many incorrect sequences were searched in NCBI.
2. EzBioCloud display sequences obtained from type strains but NCBI displayed all kinds of sequences uploaded by users and these include mis-identified matters also. Hence, if you want exact identification, should click "type materials only" box in NCBI.
2. EzBioCloud uncultured types for better classification. The first 98.99% was unclutured one and this means your strain is novel species.
Kae Kyoung Kwon
1. May I ask When I should click "type materials only" ? or should I do this every time I want to identify the sample?
2. I just changed it and it turned out the best hit was at 98.66% similarity!!
3. I am also wondering how do I know if these sequences are misidentified? In the original result (without clicking the type materials only) the top 2 hits were new isolates from one paper that were published just last year. so I'm not sure how do I determine if this is reliable or not even when this sequence result was from one of the published papers?
4. In terms of EzBioClouds, how do you know it is an uncultured type? is it because it doesn't have a taxon name?
Also circling back to Blast, the hit strain name of the first hit on EzbioClouds happen to be the first hit on Blast when I clicked "type materials only". Does this mean this one has been cultured but EzbioClouds has not updated its database?
I quickly searched for the name of this first hit and it was only recently been proposed to a new species. So I guess that is why. But please let me know if anything I say doesn't make sense. Thank you!
- Badges
- Science method