I am dealing with some mined DNA sequences from GenBank of flowering plants (rbcL and matK gene), to construct phylogenetic tree out of them for finding better taxonomic resolution.
These sequences are of varied length from 250 to 2500 bp, possess gaps and ambiguous bases. I tried with MEGA and MAFFT server to perform alignment, but found it is not that easy to get them aligned, as both those methods failed to find either the region of similarity in matK (possess gaps) and were unsuccessful to even load rbcL gene (>1000 sequences with ambiguous bases).
Should I go with alignment free phylogeny? Kindly, recommend the method (Program/package/software/webserver) suited for the analysis...
Hi Rahul, did you already tried the SATe? It has a better results than MUSCLE, CLUSTAL and other algorithms.
http://phylo.bio.ku.edu/software/sate/sate.html
Good Luck
Jalview is a multiple sequence aligment plataform were different algoritm for MSA are included
http://www.jalview.org/
Thank you @ Wilson and Obdulia ... SATe gave an error 'Unknown Characters'. As, the input data file is mined from Genbank it might possess ambiguous characters. It is impossible to remove those as the dataset comprises of >1000 sequences
While, Jalview was not even able to load the sequence file (FASTA)
Rahul, you can try to fix it by two ways. If you have some affinity with Perl or any other language which handle Regex, you can find anything different from ACTG or N and change it. Sorry for nonot being more specific, but I'm not sure about what kind of bases SATe is able to read. Please, take a look at IUPAC codes and SATe manual. A second way could be using DNAsp software. They have a version to handle big datasets and it can import and export several different tipes of format as well as correct those different bases.
Cheers
Hi Rahul
Pl find an interesting article matching your queries
Chapter 4: Principles and Methods of Sequence Analysis
https://www.ncbi.nlm.nih.gov/books/NBK20261/
Dear Rahul,
A reference implementation of PICS-Ord is available from http://scit.us/projects/ngila/wiki/PICS-Ord/. It requires the statistical software package R and the alignment software, Ngila . It is flexible but defaults to using Ngila's zeta cost scores to construct distance matrices, and R's cmdscale function to ordinate these matrices. It supports nucleic-acid and protein data and can produce encodings using up to 64 characters. It integrates well with RAxML Version 8. It is a tool for Phylogenetic Analysis and Post-Analysis of Large Phylogenies.
try it.
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Bioinformatics
- Bioinformatic Tools