SEQanswers

SEQanswers (http://seqanswers.com/forums/index.php)
-   Bioinformatics (http://seqanswers.com/forums/forumdisplay.php?f=18)
-   -   CAP3 - strange behaviour, never finishes running (2xXeon 2620v4/Ubuntu 16.04.2 LTS) (http://seqanswers.com/forums/showthread.php?t=75656)

Timofey Skvortsov 04-25-2017 01:03 AM

CAP3 - strange behaviour, never finishes running (2xXeon 2620v4/Ubuntu 16.04.2 LTS)
 
I am running CAP3 under Ubuntu 16.04.2 LTS (2xXeon 2620v4) to assemble a very small data set (31 contigs in 7k-700k bp range). I am doing that to check whether or not the contigs can be merged by overlap. CAP3 loads one thread 100%, reserves 200MB RAM and sits like that for ages, never finishes running, never fails.

The same data set gets processed by CAP3 in approximately 15 seconds on my Macbook with Intel i7 4th gen and MacOS Sierra.

The CAP3 versions I use are 64bit for Intel Linux (new, not the 2009 version) and Mac, downloaded from http://seq.cs.iastate.edu/cap3.html

Has anyone seen anything like that before? On a side note, if anyone could recommend a good alternative to CAP3 for merging contigs, I would be very grateful.

Brian Bushnell 04-25-2017 11:56 AM

Minimus2 can also be used for that purpose. BBMap's Dedupe can detect and print out the overlaps so you can view them as a graph, but does not actually merge anything.

Timofey Skvortsov 04-25-2017 12:48 PM

Thanks, Brian. I tried using Minimus2 before, still have a feeling that CAP3 produces better results. As I am pretty sure that my contigs don't have any overlaps, might give Dedupe a go.


All times are GMT -8. The time now is 04:53 AM.

Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2020, vBulletin Solutions, Inc.