chain alignment track of multiple genomes

[Dominik Hadžega]

Dear people at UCSC,

I am using ucsc browser for a study with cow genome. We would like to view multiple chain alignment of at least 8 different mammals on cow genome (something like 20 mammals track for hg38) to see conservation between sequences. Do you have any advise how can I properly add this kind of alignement to custom tracks and what kind of alignment file should I produce?

Thank you very much in advance,

Dominik Hadžega

[Daniel Schmelter]

Hello Dominik,

Thank you for writing to the Genome Browser with your question about multiple alignments to the cow genome.

Running the whole genome alignments can be a somewhat complicated process. The most complete documentation on the topic can be found here:

http://genomewiki.ucsc.edu/index.php/Whole_genome_alignment_howto

I will explain the steps briefly here. To summarize

1. Run each pair-wise alignment with the doBlastzChainNet.pl script, producing pairwise MAF files
2. Collect those pairwise-MAF files in one place, run MultiZ to produce a multi-species MAF file
3. Visualize that multi-species MAF file on the Genome Browser

First, you will need pair-wise alignment MAF files between each organism and the cow assembly you are using. We have .axt files for select organisms in the bosTau8/vsGenome directories (http://hgdownload.soe.ucsc.edu/goldenPath/bosTau8/) which can be converted to the MAF input format for MultiZ using the axtToMaf utility available in our utilities directory (http://hgdownload.soe.ucsc.edu/admin/exe/linux.x86_64/). Since you may want to create your own chain files, you will need to use the program doBlatzChainsNets.pl, which requires 2bit and chrom.sizes input files. Generating these will require two additional utilities "faToTwoBit" and "faSizes -detailed". The pairwise MAF generation process is described in detail here:

http://genomewiki.ucsc.edu/index.php/DoBlastzChainNet.pl

This will also generate chain files that can be viewed as bigChain files (https://genome.ucsc.edu/goldenPath/help/bigChain.html). There are three different types of pairwise MAFs generated with different filtering methods depending on the sensitivity you want.

• Full - keeps all of the alignments for highest coverage, for distant relatives
• Reciprocal best - highest standards, where query and target are both the same best match, for moderate distance relatives or poorly assembled genome.
• Syntenic - maximizes connections, evolutionarily based for close relatives with well-assembled genomes

Once you have your pairwise MAF files made, you will need to run the MultiZ program to generate multi-species MAF files. For a reference on how the human 20-way MultiZ alignment was made, you can refer to the makeDoc log of the process:

https://genome-source.gi.ucsc.edu/gitlist/kent.git/blob/master/src/hg/makeDb/doc/hg38/multiz20way.txt

Then you will have your multi-species MAF files, which can be converted to bigMaf and displayed on the Genome Browser. You can read more about MAF and bigMaf formats here:

https://genome.ucsc.edu/FAQ/FAQformat.html#format5
https://genome.ucsc.edu/goldenPath/help/bigMaf.html

I hope this was helpful. If you have any more questions, please reply to gen...@soe.ucsc.edu. All messages sent to that address are publicly archived. If your question includes sensitive data, you may send it instead to genom...@soe.ucsc.edu.

All the best,
Daniel Schmelter
UCSC Genomics Institute

--

---
You received this message because you are subscribed to the Google Groups "UCSC Genome Browser Public Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email to genome+un...@soe.ucsc.edu.
To view this discussion on the web visit https://groups.google.com/a/soe.ucsc.edu/d/msgid/genome/CAGCSuWZitDGZy-L4BoT%3DgmfF5fjGD4riZM4KNN__9RmNUcia-w%40mail.gmail.com.