SEA-PHAGES | % Identity vs % Aligned vs % Coverage

Link to this post \| posted yesterday, 02:56
storksle	Hi all, Is there a good resource that describe the difference between % Identity vs % Aligned vs % Coverage? These terms are all used in the CDD output on PECAAN and I would like to be able to differentiate them easier. I am getting some snippets here and there on my searches, but nothing concrete. Any help?

Link to this post \| posted yesterday, 12:15
debbie	Have you looked here? https://blast.ncbi.nlm.nih.gov/doc/blast-help/ There is a glossary of terms included there. Best, debbie

Link to this post \| posted today, 16:56
cdshaffer	Here is a simple example with super simple sequences: `query = AAA subject = CCCAAATTTGGG` When blast does the alignment with appropriate settings (since blast would never show this by default as it is too short, but you can force it) and you would in theory get this result: `1 AAA 3 \|\|\| 4 AAA 6` The %aligned is 100% because the entire query is found in the alignment the % coverage is 25% since only 3 bases of the 12 bases in the subject are in the alignment For % identity you get 100 % because 100% of the bases in the alignment match identically For DNA there is no %similar (the % similar is only used for amino acids alignment) but for a.a you just would count the fraction of alignment columns that are either identical or similar and divide by the length of the alignment. CCD is a database of protein domains (i.e. small parts of proteins seen widely) think things like zinc finger or ATP binding domain. Thus, for interpretation of CCD hits hits you care most about the % coverage and % similar the % aligned is mostly irrelevant.

Link to this post | posted today, 16:56

cdshaffer

Here is a simple example with super simple sequences:

query = AAA

subject = CCCAAATTTGGG

When blast does the alignment with appropriate settings (since blast would never show this by default as it is too short, but you can force it) and you would in theory get this result:


1 AAA 3
  |||
4 AAA 6

The %aligned is 100% because the entire query is found in the alignment
the % coverage is 25% since only 3 bases of the 12 bases in the subject are in the alignment
For % identity you get 100 % because 100% of the bases in the alignment match
identically
For DNA there is no %similar (the % similar is only used for amino acids alignment)
but for a.a you just would count the fraction of alignment columns that are either identical or similar and divide by the length of the alignment.

CCD is a database of protein domains (i.e. small parts of proteins seen widely) think things like zinc finger or ATP binding domain.
Thus, for interpretation of CCD hits hits you care most about the % coverage and % similar
the % aligned is mostly irrelevant.

Recent Activity

% Identity vs % Aligned vs % Coverage