I am trying to work on a pipeline to generate optimize Multiple Sequence Alignments (MSAs) based on NCBI pulls. While I am doing this I am comparing how the length of alignments, consensus sequence, and pairwise % change and improve with each step.
It has lead to a question that I can't find a good reference to in the literature. What would you consider a good pairwise % for a MSA? I was thinking over 50%?
Open to thoughts and what you think are other benchmarks for a good and strong MSA
•
written
by
aframirez •
1 answer
There is no pairwise alignment in MSA, because pairwise alignment by definition means that a pair of sequences is involved. There are many pairwise alignments between sequences in MSA - I assume this is what you are asking about.
If by good you mean a reliable alignment, over 50% identity may be a good target. If by good you mean an alignment between related sequences, there will be many related sequences from evolutionary distant organisms with pairwise alignment identities <50%, and probably all the way down to 30% would be reasonable.
Log in to answer this question.
-
Can you force two sequences to align with pairwise alignment?
written by msening •
-
How do I quickly run through all possible merger outputs?
written by lzbenicmi •
-
How to obtain multiple sequence alignments of genomes of selected vertebrate species?
written by kousi31 •
-
Using a Multiple Seq Alignment tool for pairwise alignments possible?
written by nchuang
-
Ideal sequence % identity for profile construction
written by Anand Rao
-
MSA building algorithm
written by ga29qal •
-
Can I Infer Conservation Scores For 7-Way Alignments With Horse Using Pairwise Alignments On Ucsc?
written by dustar1986
-
Loop Thousands Of Orthologous Msas Into Raxml
written by Louis •
-
Verifying A Quality Improvement For Multiple Sequence Aligments (Msa)
written by Aleksandr Levchuk
-
Novel Strategy For Improving Sequence Alignment
written by Pals
