how to measure similarity from EMBOSS Needleman output

Question

[nobu.kim66]

It says,

The Identity: is the percentage of identical matches between the two sequences over the reported aligned region (including any gaps in the length).

The Similarity: is the percentage of matches between the two sequences over the reported aligned region (including any gaps in the length).

here: http://embossgui.sourceforge.net/demo/manual/needle.html

So it's percentage of matches v. percentage of identical matches.

What is the difference?

Answer 1

[Jean-Karim Heriche]

A match is defined as two aligned residues with a score above a threshold (here threshold is 1, they are indicated by : in the output, see the markup line description) so they can be different but having a high score means they are considered similar. Identity is only concerned with whether the two aligned residues are exactly the same regardless of score.

Answer 2

[nobu.kim66]

Thanks! That really clarifies it for me.