A suffix array can be a useful data structure for representing sequence content. MUMmer is perhaps the best-known tool to use suffix arrays, but the speed and capacity of MUMmer can be improved upon, given some trade-offs. sparseMEM offers a drop-in replacement for the mummer program that is reported to be more memory-efficient than that in MUMmer, with a couple of differences:
- Economy of memory, at the cost of greater run time
- Multithreading of query matches
As I'm trying to align pairs of Very Large Genomes, I thought I would give MUMmer or something similar a try. The mummer program seems to be working on the genomes, though it is necessary to make it 64-bit compatible by following the advice here, by adding a line before src/kurtz/libbasedir/types.h so that lines 48-49 look like
#define SIXTYFOURBITS
#ifdef SIXTYFOURBITS
Kudos to MUMmer's developers for taking this into account. I would still like to try a more memory-efficient alignment, especially one that offers multithreading. Hence my use of sparseMEM.
BUT ... and here's the reason for this repository ... sparseMEM won't work "out of the box." It fails for the large (~20Gbp) genomes I am trying to align.
NOTE: it still fails, even with the changes contained here... I have no explicit message for the latest failure beyond "Killed". Judging from htop what may have happened is that it wanted too much memory, last I checked it was using close to 1TB on a 1TB machine.
The changes required to get sparseMEM to work for such large genomes are quite a bit more extensive than for the MUMmer suite, but no more complicated. They are easily compartmentalised by using typedef similar to MUMmer, so that's what I've done in the code available here.
Another change from sparseMEM is that the built executable is named mummer-sparseMEM-big, rather than mummer as in the original sparseMEM release.
This executable has support for large genomes enabled by default. You can disable this in the Makefile by removing the -DSIXTYFOURBITS flag from FLAGS. Note that the name of the executable will remain as above.
I haven't tried any of these tools, but maybe you will.
essaMEM by Vyverman et al. offers further memory economy.
backwardMEM by Ohlebusch et al. is reported by Vyverman et al. to outperform sparseMEM unless abundant memory is available, but the developers note at the backwardMEM download site that it is not currently capable of handling genomes > 2Gbp.
vmatch is closed-source.
sparseMEM was developed by Khan et al. My modifications are only to enable sparseMEM to work on large genomes. The original license of sparseMEM is found in the file LICENSE in this repository, as well as in the sparseMEM zip file at the developers' download site, and reads in full:
* Copyright (c) 2009, Princeton University * All rights reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions are met: * * Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * * Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * * Neither the name of the Princeton University nor the * names of its contributors may be used to endorse or promote products * derived from this software without specific prior written permission. * * THIS SOFTWARE IS PROVIDED BY Princeton University ''AS IS'' AND ANY * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE * DISCLAIMED. IN NO EVENT SHALL Princeton University BE LIABLE FOR ANY * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
Consider a similar license to apply to these modifications, with this added copyright applying to my changes only.
* Copyright (c) 2013, Douglas G. Scofield, [email protected]