Juicer

Please make sure you run this from login node, or submit it as sbatch job.
If you run it under interactive job, the srun command inside the script will not wait for the split jobs to finish and start alignment jobs before split, that will cause problems.
# login to O2
ssh youraccount@o2.hms.harvard.edu




# make a folder in scratch 
mkdir -p /n/scratch/users/a/abc123/juicerTest/fastq


# go to the folder
cd /n/scratch/users/a/abc123/juicerTest


# copy over some testing data
cp /n/shared_db/misc/juicerData/fastq/* fastq/


# run the workflow
/n/shared_db/misc/juicer/juicer.short.2hours.2core.sh -g mm10 -d $PWD -s MboI -p /n/shared_db/misc/juicerReference/mm10.chrom.sizes -y /n/shared_db/misc/juicerReference/mm10_MboI.txt -z /n/shared_db/misc/juicerReference/Mus_musculus_assembly10.fasta -D /n/shared_db/misc/juicer/SLURM

#output:
(-: Looking for fastq files...fastq files exist
(-: Aligning files matching /n/scratch/users/a/abc123/juicerTest/fastq/*_R*.fastq*
 in queue commons to genome mm10 with site file /n/shared_db/misc/juicerReference/mm10_MboI.txt
(-: Created /n/scratch/users/a/abc123/juicerTest/splits and /n/scratch/users/a/abc123/juicerTest/aligned.
srun: job 13037522 queued and waiting for resources


# wait here for a while until the split job finishes
srun: job 13037522 queued and waiting for resources
srun: job 13037522 has been allocated resources
list of splitted fastq files:
total 1185828
-rw-rw-r--. 1 ld32 ld32 129231923 Apr 27 12:00 HIC003_S2_L001_R1_001.fastq000.fastq
-rw-rw-r--. 1 ld32 ld32 129234906 Apr 27 12:00 HIC003_S2_L001_R1_001.fastq001.fastq
-rw-rw-r--. 1 ld32 ld32 129255288 Apr 27 12:00 HIC003_S2_L001_R1_001.fastq002.fastq
-rw-rw-r--. 1 ld32 ld32 129266976 Apr 27 12:00 HIC003_S2_L001_R1_001.fastq003.fastq
-rw-rw-r--. 1 ld32 ld32  90124479 Apr 27 12:00 HIC003_S2_L001_R1_001.fastq004.fastq
-rw-rw-r--. 1 ld32 ld32 129231923 Apr 27 12:00 HIC003_S2_L001_R2_001.fastq000.fastq
-rw-rw-r--. 1 ld32 ld32 129234906 Apr 27 12:00 HIC003_S2_L001_R2_001.fastq001.fastq
-rw-rw-r--. 1 ld32 ld32 129255288 Apr 27 12:00 HIC003_S2_L001_R2_001.fastq002.fastq
-rw-rw-r--. 1 ld32 ld32 129266976 Apr 27 12:00 HIC003_S2_L001_R2_001.fastq003.fastq
-rw-rw-r--. 1 ld32 ld32  90124479 Apr 27 12:00 HIC003_S2_L001_R2_001.fastq004.fastq


# for large data set, please use
/n/shared_db/misc/juicer/juicer.short.12hours.4core.sh ...

# for troubleshooting, look log files in:
ls -l debug
# to run hiccups: 
module load gcc/4.8.5 java/jdk-1.8u112 cuda/8.0 
sbatch -p gpu --gres=gpu:1 -t 120 --mem 3G --wrap "java -Xmx2G -jar /n/shared_db/misc/juicer/SLURM/scripts/juicer_tools.1.7.6_jcuda.0.8.jar hiccups -m 500 -r 5000,10000 -f 0.1,0.1 -p 4,2 -i 7,5 -d 20000,20000 --ignore_sparsity $PWD/aligned/inter.hic $PWD/aligned/hiccups" 

# this command will create result in: /n/scratch/users/a/abc123/juicerTest/hiccups