Sign inSign up

samesense/run-proseq

By samesense

•Updated about 8 years ago

Dockerfile w/ pre-build run-proseq env and singularity.

Image
0

10K+

samesense/run-proseq repository overview

⁠run-proseq

  • mm9 mapped w/ novoalign or bwa, and optional normalization by whole fly spike-in
  • While a bam file is the best description of activity, here a special bigwig file is made that just contains one base at the end of each read. These bigwig files are used for all calculations, because using the full read (and its pileup) is incorect.
  • Sample names⁠
  • Sample email⁠
  • Methods⁠

⁠Pipeline installation

If you can install Singularity⁠, follow Linux instructions. If not, follow Docker instructions.

⁠Data references
  • use rule or igenomes
⁠Linux install

Install Singularity⁠, Miniconda⁠, and Git⁠.

# clone this repo
git clone https://github.com/samesense/run-proseq/
cd run-proseq

# create anaconda enviroment
conda env create -f requirements.txt

# on respublica, load singularity module
module load singularity
⁠Docker install

Install Docker⁠, then

docker pull samesense/run-proseq

⁠Pipeline running

All work is done inside your cloned directory (run-proseq) using an anaconda enviroment.

⁠Docker run
docker run -it -v /absolute/local/path/:/external/ --privileged=True --cap-add=SYS_ADMIN --detach-keys="ctrl-@" samesense/run-proseq

# clone repo
cd /absolute/local/data/
git clone https://github.com/samesense/run-proseq/

# navigate to your working directory (external)
cd /external/run-proseq

# mv sample data to correct location
mkdir data/raw/fq_gz
mv data/sample/sample*_* data/raw/fq_gz/

# activate anaconda env
source activate proseq

# make bigwig files normalized to millions of mapped reads
snakemake --configfile configs/docker_sample.json \
--use-singularity --singularity-args "-B /external/:/external/" \
-s src/rules/Snakefile.py all_bw_totalReadNorm
⁠Linux run
# activate anaconda enviroment
source activate proseq
⁠Updating the config⁠

You need to update data paths in configs/snakemake.json⁠. Mention normalization group.

⁠Samples

Put your fq files (sampleName_Read1.fq.gz and sampleName_Read2.fq.gz) into `run-proseq/data/raw/fq_gz/. Add your sample name to the SAMPLES list in configs/snakemake.json⁠.

⁠Running Snakefile.py
# run fastqc 
snakemake --configfile configs/snakemake.json \
--use-singularity --singularity-args "-B /mnt/isilon/:/mnt/isilon/" \
-s src/rules/Snakefile.py all_fastqc

# make bigwig files normalized to millions of mapped reads
snakemake --configfile configs/snakemake.json \
--use-singularity --singularity-args "-B /mnt/isilon/:/mnt/isilon/" \
-s src/rules/Snakefile.py all_bw_totalReadNorm

# make bigwig files normalized by fly and millions of mapped reads
snakemake --configfile configs/snakemake.json \
--use-singularity --singularity-args "-B /mnt/isilon/:/mnt/isilon/" \
-s src/rules/Snakefile.py all_bw

# map paired-end mouse fastq files
snakemake --configfile configs/snakemake.json \
--use-singularity --singularity-args "-B /mnt/isilon/:/mnt/isilon/" \
-s src/rules/Snakefile.py all_mouse_mapping

# compute transcript pausing indices and transcript activity
snakemake --configfile configs/snakemake.json \
--use-singularity --singularity-args "-B /mnt/isilon/:/mnt/isilon/" \
-s src/rules/Snakefile.py all_transcript_stats

# make TSS plot for transcripts
snakemake --configfile configs/snakemake.json \
--use-singularity --singularity-args "-B /mnt/isilon/:/mnt/isilon/" \
-s src/rules/Snakefile.py all_tss_plots

Replace /mnt/isilon/ with your base path.

⁠Output files
⁠Bigwig files
  • Bigwig files are stored in sub directories of {DATA}processed/mm9/.
  • The pileup signal is stored in two types of bigwig files. Read bigwig files describe the pileup of the full read. Base bigwig files consider only one end of each read then makeing the pileup. This way, each read is only counted once in each bigwig file.
  • Bigwig files come in three normalization flavors:
    • un-normalized: bigwig_sort/
    • normalized by fly spike-in: bigwig_sort_scaleFly/
    • normalized by total reads scaled by 1 million reads: bigwig_sort_scaleTotReads/
⁠Summary stats files
  • {DATA}processed/{group}.{tool}.final_mapped_counts.csv: Mapped read counts organized by noramlizatoin group and alignment tool.
  • {DATA}processed/{group}.{tool}.fly_normalization.csv: Scaling factors determined by deseq2. Also organized by normalization group and alignmetn tool.

Tag summary

Content type

Image

Digest

Size

3 GB

Last updated

about 8 years ago

docker pull samesense/run-proseq