The sequence changes with the refseq version

Question

[jinkuozhang]

I want to know why and how NCBI update the version of accession number. Like this one: from NM_004764.3 to NM_004764.4: three bases were changed.

Thanks!

[jinkuozhang]

finswimmer, Thank you!

I have NM_004764 amplified from different cDNA libraries, there sequence are the same as NM_004764.3. Now I know NCBI make minor correction to update the refeseq. However what is its Criterion that version 3 is "wrong" and version 4 is "correct"?

[GenoMax]

It is difficult to know for sure what evidence NCBI used to make the correction but here is what the notes for that record say:

REFSEQ: This record has been curated by NCBI staff. The
            reference sequence was derived from DB072454.1, BC028581.2,
            AF104260.2 and DB527532.1.
            On Jul 8, 2010 this sequence version replaced NM_004764.3.

So the sequence has been there since 2010. I don't think either version is wrong. The latest version represents the current information NCBI (someone who looked at the evidence) has about this sequence.

Note: Please use ADD COMMENT/ADD REPLY buttons when responding to existing posts to keep threads logically organized.

[jinkuozhang]

genomax, Very nice information. Can I look the difference between NM_004764.3 and NM_004764.4 as SNPs?

[GenoMax]

I am not sure what to tell you. You have experimental evidence that seems to support the older version of the sequence (so that is hard to refute) but at the same time RefSeq sequence seems to be unchallenged/unchanged since 2010.

Take a look at the gene record for PIWIL1 and see if you are able to glean any additional information.

Are those three differences randomly scattered across the length?

Answer 1

[finswimmer]

Hello,

NCBI describes this in their FAQ.

fin swimmer