I have 16S gene sequences and some of them are quite different. When I create a tree using mega I can see that they are closely related but not very close? Whats the criteria for classifying bacteria to species level?
Thanks in advance.
Bruno... the accepted limit of 97% of similarity to define an OTU is redefined to 98.7% (see Stackenbrand 2006 and Yarsa 2014). Yo can use Silva database (online) and classified your sequence with the SINA tool. If you want a nice tree, ARB (run in linux) software is excelent, otherwise, use the SINA aligment and construct the tree with MEGA using a NJ-tree algorithm.
Best.
Hallo..
to my knowledge, if you want to build a good tree, you should add a reference species that you can get from NBCI. 10 from the same species, 5 from the same genus and 1 outgroup..
Do you mean you want to identify until species level? have you try to BLAST it using NCBI?
Ira
I have 16S genes from different species and a few of the 16S genes are not in blast as we have whole genome sequenced some bacteria and they were previously thought to be the ones which we wanted to sequence but it turned out that they are different......When we took the 16S genes from this bacteria(and the known ones) and drew trees then they were associated but not exactly the same as to the species that have been previously discovered.
I want to find out if the 16S genes fall into a different category of species that is previously unknown so that we can name is.
I want to know how disimilar the 16S sequences should be so as to classify the organism at the species level.
thanks, Bruno
Usually, for pyrosequencing, people use a limit of 97% of similarity to define an OTU. But his limit is often subject to discussion in the literature. To my knowledge ther is no clear consensus on this value. What % of similarity do you have between your sequences?
You can also do a comparison based on more genes (16S, RpoB, GyrA etc..) specially if you have the whole genome.
first, the 16S sequence quality should be checked. U should sequence the region at-least twice to confirm the reproducibility. Even small difference in the base pair that is not in the variable region, the target species showed different phylogenetic position. better, plz refer IJSEM journal. u can find out the answer you want.
Bruno... the accepted limit of 97% of similarity to define an OTU is redefined to 98.7% (see Stackenbrand 2006 and Yarsa 2014). Yo can use Silva database (online) and classified your sequence with the SINA tool. If you want a nice tree, ARB (run in linux) software is excelent, otherwise, use the SINA aligment and construct the tree with MEGA using a NJ-tree algorithm.
Best.
Hi Alejandro, thanks for your response! Can you please give me the full references of the papers you mentioned? thanks!!
http://www.nature.com/nrmicro/journal/v12/n9/full/nrmicro3330.html
If the 16S rRNA gene sequence similarity of your strain is less than 97% with the other validly published strains, then your strain is promising novel. But it not necessary. If the value is more than 97% then DNA-DNA is performed. In many articles bacterial strains are validated as novel even the sequence similarity is more than 97%. here i am attaching one article. kindly find the attachment.
- Badges
- Science topic
- Similar topics
- Biological Science
- Microbiology