MUMmer's (probably) strange behavior

[flobpf]

Hi,

I'm trying to use MUMmer to align two plant genomes - A. thaliana (120MB) and A. lyrata (240MB). The two species are fairly close - 5 million years - similar to humans and chimps.

I used nucmer with the following command line to find alignment-blocks

 nucmer -c 100 --prefix=ATref_ALquery AT.fasta AL.fasta

Then, I parsed the delta file to reveal what percentage of each of the query scaffolds were covered on the reference genome. I was expecting to find a large percentage (maybe >70%) of each AL scaffold to map to one AT chromosome and the remaining 30% to be rearranged between various other chromosomes.

However, what I find is for each AL scaffold, only 30-60% of the scaffold has any kind of alignment. For example, for the largest scaffold in A. lyrata, only 35.2% maps to Chromosome1 of A. thaliana, while another 3% maps to all other chromosomes (only 38% of the scaffold has any sort of alignment).

Wonder if I'm doing something wrong or whether this is a standard observation for MUMmer runs? I'd appreciate any suggestions on how to improve the percentage of alignment.

Thanks

UPDATE: Just to be sure, I also performed an AT vs AT search and the results are as expected - 100% mapping of each AT chromosome.

[seb567]

Maybe there is an option to tell MUMmer to process both strands and you forgot it.

Hi,

I'm trying to use MUMmer to align two plant genomes - A. thaliana (120MB) and A. lyrata (240MB). The two species are fairly close - 5 million years - similar to humans and chimps.

I used nucmer with the following command line to find alignment-blocks

 nucmer -c 100 --prefix=ATref_ALquery AT.fasta AL.fasta

Then, I parsed the delta file to reveal what percentage of each of the query scaffolds were covered on the reference genome. I was expecting to find a large percentage (maybe >70%) of each AL scaffold to map to one AT chromosome and the remaining 30% to be rearranged between various other chromosomes.

However, what I find is for each AL scaffold, only 30-60% of the scaffold has any kind of alignment. For example, for the largest scaffold in A. lyrata, only 35.2% maps to Chromosome1 of A. thaliana, while another 3% maps to all other chromosomes (only 38% of the scaffold has any sort of alignment).

Wonder if I'm doing something wrong or whether this is a standard observation for MUMmer runs? I'd appreciate any suggestions on how to improve the percentage of alignment.

Thanks

UPDATE: Just to be sure, I also performed an AT vs AT search and the results are as expected - 100% mapping of each AT chromosome.

[Tong.W]

You need not set -c for 100,maybe it is too large.Or you may try to promer instead.