JumpHash: Error writing output

[GoogleCodeExporter]

What steps will reproduce the problem?
1. >MosaikJump -ia ../../NGS_Programs/1kgdata/ref/human_g1k_v37_mosaik.dat -out 
../../NGS_Programs/1kgdata/ref/human_g1k_v37_mosaik_15jmp -hs 15

The input is the 1000 genomes reference.
------------------------------------------------------------------------------
MosaikJump 1.1.0021                                                 2010-11-10
Michael Stromberg                 Marth Lab, Boston College Biology Department
------------------------------------------------------------------------------

- retrieving reference sequence... finished.

- hashing reference sequence:
100%[===========================================================================
==================] 1,679,606 hashes/s       in 30:46  

- serializing final sorting vector... finished.

- writing jump positions database:
ERROR: Tried to store an empty hash.
2.
3.

What is the expected output? What do you see instead?

What version of the product are you using? On what operating system?
MosaikJump 1.1.0021     OpenSuSe x64 Linux

Please provide any additional information below.

Original issue reported on code.google.com by 2007...@gmail.com on 15 Dec 2010 at 2:29

[GoogleCodeExporter]

Tried this with other sequences as well such as chr1.fa and still get the same 
error
ERROR: Tried to store an empty hash.

Original comment by 2007...@gmail.com on 15 Dec 2010 at 6:11

[GoogleCodeExporter]

Hi there,

I guess the problem is reading temporary files fails.
Is that human genome sequences? If it is, I remember the space request may be 
up to 50G.

Could you please check the free disk space?

Thanks,
Wan-Ping

Original comment by WanPing....@gmail.com on 24 Dec 2010 at 4:46

• Changed state: Accepted

[GoogleCodeExporter]

I have around 300Gb free on the disk.

I tried this with one of the Hapmap project files, and also my own data..
however this is happening with both data.

 On Dec 24, 2010 11:47 AM, <mosaik-aligner@googlecode.com> wrote:

Original comment by 2007...@gmail.com on 25 Dec 2010 at 12:59

[GoogleCodeExporter]

Hi there,

Could you please change MOSAIK_TMP under your directory?

> mkdir ~/tmp
> export MOSAIK_TMP=~/tmp

Original comment by WanPing....@gmail.com on 28 Dec 2010 at 1:49

[GoogleCodeExporter]

Seems to have sorted it out!

- retrieving reference sequence... finished.

- hashing reference sequence:
100%[=======================================================] 1,619,007
hashes/s       in 31:55

- serializing final sorting vector... finished.

- creating jump keys database (273 blocks):
blocks: 273 (0.1531 blocks/s)

- writing jump positions database:
100%[===============================================]  97,483.9 hash
positions/s   in   8:09:47

MosaikJump CPU time: 18080.910 s, wall time: 33224.760 s

Thank you WanPing

Original comment by 2007...@gmail.com on 28 Dec 2010 at 1:26

[GoogleCodeExporter]

Unfortunately, the problem just got shifted..
When using MosaikAlign, I now get the error mentioned in Issue #25
 Using the following alignment algorithm: all positions
- Using the following alignment mode: aligning reads to all possible locations
- Using a maximum mismatch threshold of 4
- Using a hash size of 17
- Using 7 processors
- Using an alignment candidate threshold of 20bp.
- Setting hash position threshold to 100
- Using a jump database for hashing. Storing keys & positions in memory.
- loading reference sequence... finished.
- loading jump key database into memory... terminate called after throwing an 
instance of 'std::bad_alloc'
  what():  St9bad_alloc

Original comment by 2007...@gmail.com on 28 Dec 2010 at 8:26

[GoogleCodeExporter]

Hi there,

How much memory you have in this machine? For the hash size 17, I think the 
memory requirement would be up to 100G.

Original comment by WanPing....@gmail.com on 28 Dec 2010 at 9:25

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

MemTotal:     66002864 kB
MemFree:       7453748 kB
Buffers:       2470336 kB
Cached:       36184592 kB
SwapCached:       5480 kB
Active:        9750544 kB
Inactive:     34868892 kB
HighTotal:           0 kB
HighFree:            0 kB
LowTotal:     66002864 kB
LowFree:       7453748 kB
SwapTotal:    18881680 kB
SwapFree:     18864164 kB
Dirty:             396 kB
Writeback:           0 kB
AnonPages:     5964200 kB
Mapped:           7544 kB
Slab:         13871732 kB
PageTables:      18076 kB
NFS_Unstable:        0 kB
Bounce:              0 kB
CommitLimit:  51883112 kB
Committed_AS:  6141980 kB
VmallocTotal: 34359738367 kB
VmallocUsed:    276816 kB
VmallocChunk: 34359461167 kB
HugePages_Total:     0
HugePages_Free:      0
HugePages_Rsvd:      0
Hugepagesize:     2048 kB

Original comment by 2007...@gmail.com on 28 Dec 2010 at 9:28

[GoogleCodeExporter]

I'm sorry for the inconvenience, but could you please try hash size 15?

Original comment by WanPing....@gmail.com on 28 Dec 2010 at 9:29

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

I have successfully completed the jump database creation and now when trying to 
align SOLiD colorspace data, I also see the Issue # 84

Original comment by 2007...@gmail.com on 30 Dec 2010 at 9:22