where to download full genome with high quality sequences?

[arkilis]

Hi all,

I am trying to find some full genomes with high quality sequences. Such as human or mouse.

I know I can download it from:

ftp://ftp.ncbi.nih.gov/genomes/

Is there any recommendation except the above one?

Many thanks!

[dpryan]

Ensembl is also good, as is JGI (you'll need to create an account there).

[blancha]

I second Ensembl.
The genome and annotation files are well organized, quick to download and complete.

You'll have to choose if you want to use the older reference genomes, or the newer reference genomes.

For the mice genome, I would recommend going with the newer mm10/GRCm38 genome. Some researchers still prefer working with mm9/GRCm37 if they have older annotation sets.

For the human genome, I'm still on the fence if one should work with the newer hg38/GRCh38 genome or the older hg19/GRCh37 genome. Ensembl has made hg38/GRCh38 the default genome, but the annotation tracks on the UCSC Genome Browser for hg38/GRCh38 are still incomplete.

[arkilis]

I second Ensembl.
The genome and annotation files are well organized, quick to download and complete.

You'll have to choose if you want to use the older reference genomes, or the newer reference genomes.

For the mice genome, I would recommend going with the newer mm10/GRCm38 genome. Some researchers still prefer working with mm9/GRCm37 if they have older annotation sets.

For the human genome, I'm still on the fence if one should work with the newer hg38/GRCh38 genome or the older hg19/GRCh37 genome. Ensembl has made hg38/GRCh38 the default genome, but the annotation tracks on the UCSC Genome Browser for hg38/GRCh38 are still incomplete.

Thanks for your suggestion. My case is quite simple, only need to blast some queries against with such a database to know what I sequenced. Thanks again!

[SylvainL]

I second Ensembl.
The genome and annotation files are well organized, quick to download and complete.

You'll have to choose if you want to use the older reference genomes, or the newer reference genomes.

For the mice genome, I would recommend going with the newer mm10/GRCm38 genome. Some researchers still prefer working with mm9/GRCm37 if they have older annotation sets.

For the human genome, I'm still on the fence if one should work with the newer hg38/GRCh38 genome or the older hg19/GRCh37 genome. Ensembl has made hg38/GRCh38 the default genome, but the annotation tracks on the UCSC Genome Browser for hg38/GRCh38 are still incomplete.

Hi, I'm actually doing RNAseq analyses in parallel on both GRCh37.71 and GRCh38.76 /from ENSEMBL) and it seems I get more mapped reads (even though the number of uniquely mapped reads is a bit lower) on GRCh38 than on GRCh37...

Concerning the annotations dataset, the gtf from ENSEMBL seems at least as complete for GRCh38 than for GRCh37...

[arkilis]

Ensembl is also good, as is JGI (you'll need to create an account there).

Thanks! That is good resource