Biowulf High Performance Computing at the NIH
KmerGenie on Biowulf

KmerGenie estimates the best k-mer length for genome de novo assembly. Given a set of reads, KmerGenie first computes the k-mer abundance histogram for many values of k. Then, for each value of k, it predicts the number of distinct genomic k-mers in the dataset, and returns the k-mer length which maximizes this number.

References:

Documentation
Important Notes

Interactive job
Interactive jobs should be used for debugging, graphics, or applications that cannot be run as batch jobs.

Allocate an interactive session and run the program. Sample session:

[teacher@biowulf ~]$ sinteractive --gres lscratch:1
salloc.exe: Pending job allocation 63311518
salloc.exe: job 63311518 queued and waiting for resources
salloc.exe: job 63311518 has been allocated resources
salloc.exe: Granted job allocation 63311518
salloc.exe: Waiting for resource configuration
salloc.exe: Nodes cn3128 are ready for job
srun: error: x11: no local DISPLAY defined, skipping
[teacher@cn3128 ~]$ cd /lscratch/$SLURM_JOB_ID
[teacher@cn3128 63311518]$ # Use GAGE data (http://gage.cbcb.umd.edu/data/index.html) as an example.
[teacher@cn3128 63311518]$ wget \
> http://gage.cbcb.umd.edu/data/Staphylococcus_aureus/Data.original/frag_{1,2}.fastq.gz \
> http://gage.cbcb.umd.edu/data/Staphylococcus_aureus/Data.original/shortjump_{1,2}.fastq.gz
--2018-03-07 17:39:23--  http://gage.cbcb.umd.edu/data/Staphylococcus_aureus/Data.original/frag_1.fastq.gz
Resolving dtn07-e0... 10.1.200.243
Connecting to dtn07-e0|10.1.200.243|:3128... connected.
Proxy request sent, awaiting response... 200 OK
Length: 64204080 (61M) [application/x-gzip]
Saving to: “frag_1.fastq.gz”

100%[===================================================================================>] 64,204,080  45.5M/s   in 1.3s    

2018-03-07 17:39:24 (45.5 MB/s) - “frag_1.fastq.gz” saved [64204080/64204080]

--2018-03-07 17:39:24--  http://gage.cbcb.umd.edu/data/Staphylococcus_aureus/Data.original/frag_2.fastq.gz
Connecting to dtn07-e0|10.1.200.243|:3128... connected.
Proxy request sent, awaiting response... 200 OK
Length: 68386818 (65M) [application/x-gzip]
Saving to: “frag_2.fastq.gz”

100%[===================================================================================>] 68,386,818  51.5M/s   in 1.3s    

2018-03-07 17:39:25 (51.5 MB/s) - “frag_2.fastq.gz” saved [68386818/68386818]

--2018-03-07 17:39:25--  http://gage.cbcb.umd.edu/data/Staphylococcus_aureus/Data.original/shortjump_1.fastq.gz
Connecting to dtn07-e0|10.1.200.243|:3128... connected.
Proxy request sent, awaiting response... 200 OK
Length: 74810444 (71M) [application/x-gzip]
Saving to: “shortjump_1.fastq.gz”

100%[===================================================================================>] 74,810,444  84.5M/s   in 0.8s    

2018-03-07 17:39:26 (84.5 MB/s) - “shortjump_1.fastq.gz” saved [74810444/74810444]

--2018-03-07 17:39:26--  http://gage.cbcb.umd.edu/data/Staphylococcus_aureus/Data.original/shortjump_2.fastq.gz
Connecting to dtn07-e0|10.1.200.243|:3128... connected.
Proxy request sent, awaiting response... 200 OK
Length: 74594473 (71M) [application/x-gzip]
Saving to: “shortjump_2.fastq.gz”

100%[===================================================================================>] 74,594,473  85.2M/s   in 0.8s    

2018-03-07 17:39:27 (85.2 MB/s) - “shortjump_2.fastq.gz” saved [74594473/74594473]

FINISHED --2018-03-07 17:39:27--
Downloaded: 4 files, 269M in 4.3s (62.7 MB/s)
[teacher@cn3128 63311518]$ ls *.fastq.gz > inputs
[teacher@cn3128 63311518]$ module load kmergenie
[+] Loading GSL 2.2.1 ...
[+] Loading Graphviz v2.38.0 ...
[+] Loading gdal 2.0 ...
[+] Loading proj 4.9.2 ...
[+] Loading gcc 4.9.1 ...
[+] Loading openmpi 1.10.0 for GCC 4.9.1
[+] Loading tcl_tk 8.6.3
[+] Loading Zlib 1.2.8 ...
[+] Loading Bzip2 1.0.6 ...
[+] Loading pcre 8.38 ...
[+] Loading liblzma 5.2.2 ...
[-] Unloading Zlib 1.2.8 ...
[+] Loading Zlib 1.2.8 ...
[-] Unloading liblzma 5.2.2 ...
[+] Loading liblzma 5.2.2 ...
[+] Loading libjpeg-turbo 1.5.1 ...
[+] Loading tiff 4.0.7 ...
[+] Loading curl 7.46.0 ...
[+] Loading boost libraries v1.65 ...
[+] Loading R 3.4.0 on cn3128
[-] Unloading Zlib 1.2.8 ...
[+] Loading Zlib 1.2.8 ...
[+] Loading kmergenie, version 1.7044...
[teacher@cn3128 63311518]$ kmergenie inputs -t2
running histogram estimation
list of reads:
frag_1.fastq.gz
frag_2.fastq.gz
shortjump_1.fastq.gz
shortjump_2.fastq.gz
Setting maximum kmer length to: 101 bp
computing histograms (from k=21 to k=101): 41 21 51 31 61 81 71 91 101 
ntCard wall-clock time over all k values: 88 seconds 
fitting model to histograms to estimate best k
could not fit histograms-k71.histo
could not fit histograms-k81.histo
estimation of the best k so far: 21
refining estimation around [15; 27], with a step of 2
running histogram estimation
list of reads:
frag_1.fastq.gz
frag_2.fastq.gz
shortjump_1.fastq.gz
shortjump_2.fastq.gz
Setting maximum kmer length to: 101 bp
computing histograms (from k=17 to k=27): 19 17 23 21 27 25 
ntCard wall-clock time over all k values: 66 seconds 
fitting model to histograms to estimate best k
could not fit histograms-k71.histo
could not fit histograms-k81.histo
table of predicted num. of genomic k-mers: histograms.dat
recommended coverage cut-off for best k: 3
best k: 19

[teacher@cn3128 63311518]$ exit
salloc.exe: Relinquishing job allocation 63311518
[teacher@biowulf ~]$

Batch job
Most jobs should be run as batch jobs.

Create a batch input file (e.g. kmergenie.sh). For example:

#!/bin/sh
set -e
module load kmergenie
test -n "$SLURM_CPUS_PER_TASK" || SLURM_CPUS_PER_TASK=2

kmergenie inputs.fofn -t $SLURM_CPUS_PER_TASK

Submit this job using the Slurm sbatch command.

sbatch [--cpus-per-task=#] [--mem=#] kmergenie.sh
Swarm of Jobs
A swarm of jobs is an easy way to submit a set of independent commands requiring identical resources.

Create a swarmfile (e.g. kmergenie.swarm). For example:

kmergenie input1.fofn -t $SLURM_CPUS_PER_TASK
kmergenie input2.fofn -t $SLURM_CPUS_PER_TASK
kmergenie input3.fofn -t $SLURM_CPUS_PER_TASK
kmergenie input4.fofn -t $SLURM_CPUS_PER_TASK

Submit this job using the swarm command.

swarm -f kmergenie.swarm -t 2 [-g #] --module kmergenie
where
-g # Number of Gigabytes of memory required for each process (1 line in the swarm command file)
-t # Number of threads/CPUs required for each process (1 line in the swarm command file).
--module kmergenie Loads the kmergenie module for each subjob in the swarm