This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Difference between various GRCH 37 releases?

I have some Genomic coordinates based on hg19 which is equal to GRCH37 and I would like to analyze these coordinates using actual genome sequences. In Ensembl, There are several releases of Grch like GRCH37.70 and GRCH37.75 etc. I am wondering what is the actual difference between all these releases of GRCH37 ?! Are the coordinates the same for all these releases?!

genome
• 10,546 views • link
updated by Gjain Gjain • written by morteza.mahmoudisaber morteza.mahmoudisaber

1 answer

This is the website of the Genome Reference Consortium, might be informative for you to poke around there

hg19 is typically thought of as a subset of GRCh37 with the GRCh37p# representing different minor / patch releases (changes do not affect coordinates)

Ensembl uses a different versioning system, you can find their list of changes here: http://www.ensembl.org/Help/ArchiveList though it seems like the latest release is mostly schema changes with some changes to annotations.

• 0 views • link
written by Ying W Ying W

Just to clarify further, versioning in Ensembl is just the annotation of the genes themselves. The primary assembly (ie coordinate system) is the same for anything that's GRCh37. A p.# will represent patches on top of the genome – things are added but the existing primary assembly is not altered in any way.

• 0 views • link
written by Emily Emily

Hi Emily, a question regarding your last sentence: how can the primary assembly coordinates remain unaltered if you ADD something? let's say in the middle of chromosome 17 there is a sequence which contains 4 repeats and not 3, as previously thought. Thus I publish a patch and now there are 4 repeats instead of only 3. I add one more repeat. Of course that changes every coordinate ... ?

• 0 views • link
written by Marvin Marvin

The three repeats are still there in the primary assembly. There's just a different version that you can (and should) instead for that locus which has four repeats.

• 0 views • link
written by Emily Emily

I don't understand this ... let's say in p6 there are 3 repeats on chromosome 17. Then in p7 there are 4 repeats. In the fasta file the sequence "> chr17" will thus be 1 repeat longer and thus all the nucleotides after that additional repeat will have a changed coordinate? You say that the three repeats are "still there" -> well of course they are still there, but there's a fourth one now, right? So the sequence becomes longer? What am I missing?

• 0 views • link
written by Marvin Marvin

Hello Marvin ,

have a look at my tutorial Which human reference genome should I use? . I tried to explain there what patches are. In short: If there is a fix/update for an existing sequence, this is not incorporated into the orginal sequence. It's getting it's own name/accession number and is included to the collection of sequences that build the reference genome.

fin swimmer

• 1 views • link
written by finswimmer finswimmer

Log in to answer this question.

More posts like this