Re: hgvs

[Reece Hart]

Hi Jianfen-

Sorry for the delay in replying. You're likely get a faster reply if you send to hgvs-d...@googlegroups.com.

First, a general warning: Genomic coordinates for transcript exon structures are generated by alignments. RefSeq, which should be considered authoritative for transcripts in RefSeq, uses Splign. UCSC uses (used?) BLAT. These aligners can and do generate different exon structures. I studied this pretty extensively several years ago. See https://www.biostars.org/p/6508/#138923 and https://www.slideshare.net/reecehart/hvp-2014-clinical-significance-of-transcript-alignment-discrepancies.

However, that's unrelated to what you're seeing. The summary is that hgvs is correct.

The clue is your own comment:

(3)the ref of hgvs is TCAGGGTGATGATGGAATT,but I give the var is  NM_033380.2:c.232-6_244delTTTCAGGGTGATGATGGAA,I think the ref is  TTTCAGGGTGATGATGGAA  

Align these sequences like so:

 

ref: TTTTTCAGGGTGATGATGGAATTCCAGG

mut:   TTTCAGGGTGATGATGGAA

hgvs:    TCAGGGTGATGATGGAATT

At a genomic sequence level, the result of deleting TTTCAGGGTGATGATGGAA is the same as deleting TCAGGGTGATGATGGAATT at a position 2 nucleotides later. The latter is a correctly normalized variant according to HGVS recommendations. 

So, hgvs is correct. Mutalyzer is misleading. I hesitate to say that Mutalyzer is incorrect or that this is a bug because your Mutalyzer screenshot does says that variant are not normalized, This is essentially declaring that some results will not be correct according to HGVS recommendations.

-Reece

On Mon, May 28, 2018 at 8:25 PM, jianfen man <jianfe...@gmail.com> wrote:

---------- Forwarded message ----------
From: jianfen man <jianfe...@gmail.com>

Date: 2018-05-14 10:11 GMT+08:00
Subject: Re: hgvs
To: Reece Hart <reec...@gmail.com>

（1）The command line program `hgvs-shell` ,see attached

（2）The hgvs version ：1.1.2

（3）Describe why I think the answer from hgvs is wrong：

First:see the last line of the picture(https://mutalyzer.nl/position-converter?assembly_name_or_alias=GRCh37&description=NM_033380.2%3Ac.232-6_244delTTTCAGGGTGATGATGGAA)

second:

The  information about NM_033380.2 come from:http://hgdownload.cse.ucsc.edu/goldenPath/hg19/database/refGene.txt.gz

and I can get :

See the highlighted lines，you can get
(1)the start c.232-6 is 107807106   ,not 107807108

(2)the end c.244 is 107807124,not 107807126

(3)the ref of hgvs is TCAGGGTGATGATGGAATT,but I give the var is  NM_033380.2:c.232-6_244delTTTCAGGGTGATGATGGAA,I think the ref is  TTTCAGGGTGATGATGGAA   

Best regards 

2018-05-13 0:47 GMT+08:00 Reece Hart <reec...@gmail.com>:

Hi Jianfen-

I'm happy to help, but I don't have enough information to assess your bug report. 

Would you please provide a simple demonstration case for the problem and all of the steps *as text*? A screenshot is not debuggable. Ideally, please use the variables defined with the command line program `hgvs-shell`. Also, please include the hgvs version (hgvs.__version__). And, please describe why you think the answer from hgvs is wrong.

Finally, please submit this either to hgvs-discuss or to github so that other members of the community can help.

Thanks,

Reece

On Fri, May 11, 2018 at 1:19 AM, jianfen man <jianfe...@gmail.com> wrote:

Dear Reece:

I am a newer for hgvs . 

I use hgvs :

var="NM_033380.2:c.232-6_244delTTTCAGGGTGATGATGGAA"

 

 But  I think  the start,end and ref is wrong for the var.

And :

start=107807106

end =107807124

ref=TTTCAGGGTGATGATGGAA

Best Regards !

Jianfen Man