Comparative Genomics - BLAT

[GenoMax]

You should align only 2 contigs that are most similar to each other. If you look closely at the PDF you posted you can probably make out which contigs are most similar to each other as pairs.

Again this is not going to help you a lot since you have 1.4M contigs.

If you don't know perl/python find a friend who can help parse the blat result file.

[milo0615]

For those larger contigs that are noted be similar by blat try using Mauve . You can get additional information from Mauve alignments: http://darlinglab.org/mauve/user-guide/files.html

Hi GenoMax,

I did use Mauve. Attached is a pdf with the Mauve results of largest 10 contigs for both assemblies (gerbera_alignment.pdf). However, Mauve does not work when I align the largest 10 contigs from assembly 1 to the whole assembly 2. I guess it is mostly designed for bacterial genome.

[GenoMax]

For those larger contigs that are noted be similar by blat try using Mauve . You can get additional information from Mauve alignments: http://darlinglab.org/mauve/user-guide/files.html

[milo0615]

@Milo: With 1.4M+ input sequences those viewer programs are not going to be useful (and your results are not in blast format either).

You can get an idea of coverage estimate by using Brian's suggestion in this thread: http://seqanswers.com/forums/showthread.php?t=44035 You will have to use the raw data for this. This suggestion is not directly related to question you asked, but may be worth while to do, since you are working with a unknown genome.

Hi GenoMax,

I took the longest 10 contigs from each assembly (>8500 bp with a max length of 120 K for assembly 1 and 104 K for assembly 2) and BLAT aligned them together. They seem to be pretty similar but I would like more in depth information about their differences. I am going to give it a try at what you suggested.

Thank you for your help and please let me know if you have any other ideas.


Thank you,

-Emilio

[GenoMax]

@Milo: With 1.4M+ input sequences those viewer programs are not going to be useful (and your results are not in blast format either).

You can get an idea of coverage estimate by using Brian's suggestion in this thread: http://seqanswers.com/forums/showthread.php?t=44035 You will have to use the raw data for this. This suggestion is not directly related to question you asked, but may be worth while to do, since you are working with a unknown genome.

[GenoMax]

Perhaps you should concentrate on largest (what is the size range of the largest contigs? Are there any that are 10kb and above?) ones first and see if they are related. That would make your search space smaller.

[milo0615]

Looks like you are a ways away from having a real assembly. You likely will need some custom scripting to get a meaningful answer since I assume your blat result file (even in PSL format) is probably pretty large.

If the blat results were in blast format and you just wanted to visualize them then http://bioinformatics.oxfordjournals...31/8/1305.full or http://www.biomedcentral.com/1471-2105/15/128 would have been useful.

Do you have a related reference genome available? What is the expected genome size for your samples?

Hi GenoMax,

I do no have a related reference genome. Therefore, I aligned both assemblies against each other. My specie is a diploid plant with a 2C DNA value estimated at 5.1 pg (about 5.0 Gb).

I am going to try the ones that you suggested. Is there a better way of comparing both assemblies? How can I get an alignment percentage coverage? Please let me know.

Thank you,

-Milo

[GenoMax]

Looks like you are a ways away from having a real assembly. You likely will need some custom scripting to get a meaningful answer since I assume your blat result file (even in PSL format) is probably pretty large.

If the blat results were in blast format and you just wanted to visualize them then http://bioinformatics.oxfordjournals...31/8/1305.full or http://www.biomedcentral.com/1471-2105/15/128 would have been useful.

Do you have a related reference genome available? What is the expected genome size for your samples?

[milo0615]

How big are these genomes? How many contigs are there in each?

Hi GenoMax,

Thank you for replying.

Assembly a is 444.5MB and it contains 1.4M contigs
Assembly b is 526.1MB and it contains 1.6M contigs

Please advice.

Thank you,

-Milo

[GenoMax]

How big are these genomes? How many contigs are there in each?

[milo0615]

Hello,

I have assembled two related species using Abyss. However, I would like to know if there is a significant difference between the two genomes. A way of doing this is by aligning both assemblies to one another. I used BLAT for the alignment but now I need help visualizing and interpreting the output.psl file. How can I get the total percentage coverage of the alignment? Lets say, assembly A covers x% of assembly B? Is there a better way of doing this? Please let me know. I would really appreciate your help.


Thank you,

-Milo