what is the difference between GRCh37 and hg19 ?

Question

[jack]

Hi all,

What is the difference between GRCh and hg? I have small RNA interaction data with genome like below and they have used hg19 for mapping the reads. I want to use GTF file in Ensemble which is based on GRCh. but I was wondering the hg and GRCh are the same? If not, where can I find the GTF file for the hg19?

chromosome  start    end      interaction ID  alignment score of target part  strand
chr2        1.9E+08  1.9E+08  human08246      32                              +
chr16       5.2E+07  5.2E+07  human04514      32                              -
chr3        1.1E+08  1.1E+08  human11115      36                              +
chrM        14360    14382    human08122      34                              +

[Pierre Lindenbaum]

see: Differences Between Reference Human Genome Assemblies From Different Sources

[jack]

But still, I couldn't figure it out whether hg and GRCh is the same or not. Could you please elaborate more?

[gandrescabrera]

Can I make a similar answer in here? Might be useful to others too. If its innapropiate take the according action to fix it, sorry.

Which is the difference between Homo_sapiens.GRCh38.91 and Homo_sapiens.GRCh38.78? Can I use a Homo_sapiens.GRCh38.91.bed as annotation to a genome referenced to Homo_sapiens.GRCh38.78?

Thanks in advance.

[Manuel Landesfeind]

The term "GRCh38" refers to the genomic assembly. The ".78" and ".91" refer to the version of gene annotation provided by Ensembl, i.e. where do introns/exons/etc start and end. To my knowledge, the genomic sequence for "GRCh38.78" and "GRCh38.91" should be identical and only the gene annotation should differ.

[morshed9642]

In GRCh37.91, what does that .91 mean?

[Pierre Lindenbaum]

see C: what is the difference between GRCh37 and hg19 ?

[morshed9642]

Hi Pierre..thanks for the link...I am sorry that I am still confused because when I google GRCh37.91, it does not show up...that ''.91'' comes along only with GRCh38. I understand that GRCh37 is the older version than GRCh38. But is that '0.91' a 'Patch'? If so, can it be used for both GRCh 37 or GRCh38?

Thanks for your help

[morshed9642]

Hi Pierre..thanks for the link...I am sorry that I am still confused because when I google GRCh37.91, it does not show up...that ''.91'' comes along only with GRCh38. I understand that GRCh37 is the older version than GRCh38. But is that '0.91' a 'Patch'? If so, can it be used for both GRCh 37 or GRCh38?

Thanks for your help

Answer 1

[Dan D]

I believe the genomic content for the two is identical, except for the mitochondrial contig.

The contig names are also different. GRCh37 names them `chr1`, `chr2`,,`chr3`, etc, while hg19 just has `1`, `2`, `3`.

Thus you can use the same GTF file for both (excluding mitochondrial, of course) if you do a simple replace operation for the contig names.

[Pierre Lindenbaum]

and some degenerate bases have been replaced by 'N' for chr3 and chrY. see: http://plindenbaum.blogspot.fr/2013/07/g1kv37-vs-hg19.html

[jack]

ok, thanks, so the genomic coordinates of the genes are the same in hg and GRCh ?

[Adam]

Although this is basically true, GRCh37 has a number of patches, which won't be included in hg19.

Answer 2

[Kevin Blighe]

More information to add (courtesy MatthewP): GRCh37 / hg19 / b37 / humanG1Kv37 - Human Reference Discrepancies

In essence: GRCh37 is identical to hg19 on the main contigs (chr1-24), but differ on chrM.

Kevin