BLAST result showing greater sequence length in %identities than that of the query sequence.

Question

[sumayya_tariq]

Please help me to understand the following BLAST result,

Query= sp|P14738|FNBA_STAA8 Fibronectin-binding protein A OS=Staphylococcus
aureus (strain NCTC 8325) GN=fnbA PE=1 SV=1

Length=1018

> gi|49484704|ref|YP_041928.1| fibronectin-binding protein precursor
[Staphylococcus aureus subsp. aureus MRSA252]
Length=965

Score = 1404 bits (3633),  Expect = 0.0, Method: Compositional matrix adjust.
Identities = 748/1022 (73%), Positives = 826/1022 (81%), Gaps = 61/1022 (6%)

Please notice the length of the query sequence i.e 1018 and the length from which the %identities is being shown i.e 1022. What I am not able to understand is where did this 1022 come from? I know it may be a naive question but please consider that I am new to this field.

[Sam]

Maybe the length is the target's length? E.g. your 1018 query are aligned to a sequence of length 1022 and 748 of the sequence are identical?

[sumayya_tariq]

But notice the length of the target mentioned which is 965.

[Sam]

Can you check if there is any indel? e.g.

Query    ACT--G
Target   ACTTTG

[Siva]

Yes, this seems to be case. I aligned the query and the subject sequences and there are 4 gaps in the query sequence in the alignment.

[sumayya_tariq]

Okay, so the 4 gaps have been added to the query sequence and therefore the length is now 1022. but how do I know if there are actually 4 gaps? Does QL--YK means 4 gaps?

[Siva]

No, there are two gaps in the QL--YK region and two other gaps are somewhere else in the query sequence. I copy pasted the portions of the alignment that contains the 4 gaps in the query sequence.

Query  361   RFSHVAFIKPNNGKTT-SVTVTGTLMKGSNQNGNQPKVRIFEYLGNNEDIAKSVYANTTD  419
             +F+HVA+IKP NG  + SVTVTG L +GSN+NG QP V+I+EY+G    + +SVYANT D
Sbjct  361   KFTHVAYIKPINGNNSDSVTVTGMLTQGSNENGTQPNVKIYEYVGVENGLPQSVYANTVD  420

Query  480   QL--YKYYYDRGYTLTWDNGLVLYSNKANGNEKNGPIIQNNKFEYKEDTIKETLTGQYDK  537
                   YYY+ GYTLTWDNGLVLYSNKANG+ K GPI+ +N FE+ ED+   +++GQYD 
Sbjct  481   NRYKTYYYYNNGYTLTWDNGLVLYSNKANGDGKYGPIVDSNNFEFSEDSGNGSISGQYDA  540

Query  778   VPQIHGQNKGNQSFEEDTEKDKPKYEHGG-NIIDIDFDSVPHIHGFNKHTEIIEEDTNKD  836
             VPQIHG NK N+  EEDT KDKP Y+ GG N +D + D++P + G N+  + IEEDT   
Sbjct  781   VPQIHGFNKHNEIIEEDTNKDKPNYQFGGHNSVDFEEDTLPKVSGQNEGQQTIEEDTTPP  840

[sumayya_tariq]

Okay, I get it now. Thank You so much Siva and Sam, that was very kind of you :)

[sumayya_tariq]

Yeah, it's there,

Query  480   QL--YKYYYDRGYTLTWDNGLVLYSNKANGNEKNGPIIQNNKFEYKEDTIKETLTGQYDK  537
                                YYY+ GYTLTWDNGLVLYSNKANG+ K GPI+ +N FE+ ED+   +++GQYD
Sbjct  481     NRYKTYYYYNNGYTLTWDNGLVLYSNKANGDGKYGPIVDSNNFEFSEDSGNGSISGQYDA  540

What does it mean?

[Siva]

It means either there are insertions in your subject sequence or deletions in your query sequence in those 4 positions. Or it could also be due to sequencing errors.

Answer 1

[Michael]

The percent identity is given with respect to the length of the aligned region in this output, not with respect to query or subject length. The alignment length can deviate from both query and subject in case of partial alignments and gapped alignments or both. In your case, there seem to be 4 gaps in the longer sequence.