I want to establish a phylogenetic relationship of our fungal species of interest along with the other fungal species belonging to the same order (Hypocreales). In order to perform this, we have done sequencing of our fungi using ITS and TEF regions and then performed a BLAST analysis in NCBI for identifying the closest members depending on similarity index. This was followed by retrieving some of the existing sequences (FASTA format) of other fungal species belonging to different families but sharing same order i.e. Hypocreales, available in NCBI. Next, we performed a multiple sequence alignment of the retrieved sequences and our own sequence of interest using MEGA X. We further removed the gaps, trimmed the extra regions and build the tree based on most suitable model using 1000 bootstrapping. However, the tree generated after the analysis is showing poor bootstrapping values in majority of the branches (some are below 20). How to improve this?
According to the 50 % majority rule, the bootstrap values less than 50 are not considered. You may please use some other softwares like IQTREE to perform the phylogenetic analysis, it may give good results.
There might not be a way to increase the bootstrap value without adding other sequence regions. Assuming your sequences are aligned correctly, poor bootstrap support means there simply aren't enough informative variable sites to support one branching pattern over another.
You don't say which tree construction method you used; e.g. neighbor-joining vs. maximum likelihood. Sometimes different tree construction methods will provide a little better resolution than others, but each method can also introduce its own biases, so it's not a good idea to go "shopping" for the strongest bootstrap values.
Try finding published sequences from papers, and also use other softwares like Mr Nair has recommended. Also i would recommend to use a good sequence alignment tools like mafft. And see if the sequences you are using from the BLAST are published or not.
usually in blast search various unpublished and ambiguous sequences also appears, be aware of them.
you can also trim the ends after sequence alignment and then concatenate both sequences.
if the problem still persists then like Mr. David has mentioned, your data is not enough you will need to add more data means more DNA sequence regions or more taxa.
Indeed the most likely explanation is the the sequences are not sufficiently phylogenetically informative.
- Badges
- Science topic
- Similar topics
- Agricultural Science
- Horticulture