mahulchak
commented
4 years ago Hi Matt, I will need more time to integrate lastz well into SVMU than my bandwidth is permitting me. I'd turn off LASTZ for now. Give me a couple of days to update the code.
On Mon, Aug 3, 2020 at 11:59 AM MattHuff notifications@github.com wrote:
I am running an analysis on two genomes - one with a pseudochromosome level assembly, the other with a scaffold level assembly. Using the provided command for lastz took several days for the program to complete, ultimately providing me with an empty file. Adding the following options to the lastz command fixed this, and I was able to get lastz output files in a reasonable timeframe:
--notransition --step=20 --nogapped
However, when I attempt to run SVMU using the output of this lastz run, I get a Segmentation fault (core dumped) error. Based on the documentation for this tool, I removed any whitespace from the input fasta files, but still got this error. As a sanity check - when the original lastz command no longer worked - I chose to take only a subsection of the input sequences and run lastz on them, which gave me output files - and used that on svmu; this provided me with results, and did not produce the Segmentation fault error. Once I saw this error, I reran this limited analysis using the new options from above, and attempting to run SVMU gave me the same Segmentation fault error.
Due to how long lastz took to run on the whole genomes without these options - and the ultimately fruitless results - I am unsure if there is another way to optimize lastz to run with these genomes. However, from my observations, it would seem as if these options are incompatible with SVMU, resulting in a Seg fault error. These commands lower the seeding sensitivity of lastz, and ignore gapped alignments; perhaps this plays some role in why it has been failing?
Do you have any recommendations for running lastz without breaking SVMU?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/mahulchak/svmu/issues/15, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABZQH2EPPR7VN44RYY6NWODR64CCZANCNFSM4PTT4M2A .
-- Mahul Chakraborty Department of Ecology and Evolutionary Biology University of California-Irvine Phone: 949 824 9559 Fax: 949 824 9559 Website: https://mahulchakraborty.wordpress.com/ Github: https://github.com/mahulchak
MattHuff
lipingfangs
peanut-nu
I am running an analysis on two genomes - one with a pseudochromosome level assembly, the other with a scaffold level assembly. Using the provided command for lastz took several days for the program to complete, ultimately providing me with an empty file. Adding the following options to the lastz command fixed this, and I was able to get lastz output files in a reasonable timeframe:
However, when I attempt to run SVMU using the output of this lastz run, I get a
Segmentation fault (core dumped)error. Based on the documentation for this tool, I removed any whitespace from the input fasta files, but still got this error. As a sanity check - when the original lastz command no longer worked - I chose to take only a subsection of the input sequences and run lastz on them, which gave me output files - and used that on svmu; this provided me with results, and did not produce theSegmentation faulterror. Once I saw this error, I reran this limited analysis using the new options from above, and attempting to run SVMU gave me the sameSegmentation faulterror.Due to how long lastz took to run on the whole genomes without these options - and the ultimately fruitless results - I am unsure if there is another way to optimize lastz to run with these genomes. However, from my observations, it would seem as if these options are incompatible with SVMU, resulting in a
Seg faulterror. These commands lower the seeding sensitivity of lastz, and ignore gapped alignments; perhaps this plays some role in why it has been failing?Do you have any recommendations for running lastz that does not prevent SVMU from completing?