Hi all,
I have a collection of nucleotide sequences corresponding to one functional domain of a hallmark gene of viruses. So in principle I am working with relatively-homogeneous functional sequences. They are similarly long (~1,200 nts).
What I need is to get a %identity matrix. I am seeing that BLAST pairwise local alignments are giving %identities that are not reliable. E.g.: a pair of sequences share 100% identity based on BLASTN all-versus-all but according to a MAFFT global alignment, they share 60% identity.
I need a tool giving reliable %identities with resolution enough to discriminate non-identical sequences.
Thank you,
Guillermo
BLAST probably la not aligning the ~1,200 nts of each sequence. According to the similarity between sequences, BLAST can consider a certain region of the sequence to give you the alignment values, while the regions with low similarity do not consider in the analyce.
In that case, I would believe that MAFFT global alignment gives you real identity values. Just keep in mind that if you align a large number of sequences in a single run, the complexity of the alignment for MAFFT will increase and the identity percentage will also change.
Regards
Use some freely available implementation of the Needleman–Wunsch algorithm. The fast, C++ version can be found in GitHub (https://mvdsman.github.io/blog/Simple-pairwise-alignment/). It reads and aligns only two sequences out of box but you can easily rewrite the pretty simple main() for reading more from the same file and producing "all vs all" in a loop instead. With 1200 b.p. the performance should be bearable on a good modern computer. But unless you just want the final score, your output (all vs all) will be pretty bulky. You probably need to think more what kind of output suits well for your project.
Victor Caña-Bozada responded to the question as "In that case, I would believe that MAFFT global alignment gives you real identity values. Just keep in mind that if you align a large number of sequences in a single run, the complexity of the alignment for MAFFT will increase and the identity percentage will also change".
I think of going against your answer and support the claim of Guillermo, D.H.
I had asked this question few weeks ago but have not yet been responded.
Suppose we have the following sequences:
Reference DNA sequence
Seq.-0 = A G C T T C A C T G A T
Variant DNA sequences
Seq.-1,1 = T T C T T C G G T G A T → Mismatches
Seq.-1,2 = - - C T T C - - T G A T → Gabs
Seq.-1,3 = - - C T T C G G T G A T → Mismatches and Gabs
Seq.-2,1 = A G C T A A C G T G A T → Mismatches
Seq.-2,2 = A G C T - - - - T G A T → Gabs
Seq.-2,3 = A G C T - - C G T G A T → Mismatches and Gabs
Seq.-3,1 = T A T A T C A C T G A T → Mismatches
Seq.-3,2 = - - - - T C A C T G A T → Gabs
Seq.-3,3 = T A - - T C A C T G A T → Mismatches and Gabs
Seq.-4,1 = A G G G G G A C T G A T → Mismatches
Seq.-4,2 = A G A A A A A C T G A T → Mismatches
RESULTS OF GLOBAL PAIRWISE COMPARISONS
With mismatches characters
% similarity (% gabs)
Seq.-0 and Seq.-1,1 = 66.7%
Seq.-0 and Seq.-2,1 = 66.7%
Seq.-0 and Seq.-3,1 = 66.7%
Seq.-0 and Seq.-4,1 = 66.7%
Seq.-0 and Seq.-4,2 = 66.7%
With gabs/indels characters
% similarity (% gabs)
Seq.-0 and Seq.-1,2 = 66.7% (33.3%)
Seq.-0 and Seq.-2,2 = 66.7% (33.3%)
Seq.-0 and Seq.-3,2 = 66.7% (33.3%)
With mismatches and Gabs characters
% similarity (% gabs)
Seq.-0 and Seq.-1,3 = 83.3% (16.7%)
Seq.-0 and Seq.-2,3 = 83.3% (16.7%)
Seq.-0 and Seq.-3,3 = 83.3% (16.7%)
Based on this problem and analysis, i guess that some of the existing algorithms for sequence comparison need a Benchmark review.
In addition, Guillermo, D.H, i recommend you to try with some alignment-free methods may you have some reliability.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Probability
- Probability Theory
- Reliability