RyogaLi / PPS

Dependencies

• python version 3.7

• Bowtie2 version 2.3.2

• samtools version 1.4

Build reference

• The pre-built reference files used for the analysis can be found in
  1. human grch37: /home/rothlab/rli/02_dev/06_pps_pipeline/fasta/human_ensembl/grch37
  2. human grch38: /home/rothlab/rli/02_dev/06_pps_pipeline/fasta/human_ensembl/grch38
  3. human 9.1: /home/rothlab/rli/02_dev/06_pps_pipeline/fasta/human_91
  4. yeast (palte specific): /home/rothlab/rli/02_dev/06_pps_pipeline/fasta/yeast_ref_all
• If you need to build new references, please make sure:
  1. Name for the reference is the same as name for the sequencing files. For example, the corresponding reference for scORFeome-HIP-05_L001.fastq.gz is scORFeome-HIP-05
  2. ID for each sequence matches the ORF-id in the summary file

Make summary file

• The summary files for human and yeast are premade before running the pipeline, the raw data can be found: /home/rothlab/rli/02_dev/06_pps_pipeline/target_orfs/human_summary.csv and /home/rothlab/rli/02_dev/06_pps_pipeline/target_orfs/yeast_summary.csv
• If you are making your own summary file, make sure you have a column with the name orf_name, which is the unique identifier for each ORF, this should also map with the sequence names in the fasta file you make. You can modify in main.py: analysisHuman or analysisYeast to select columns you want to keep

Input FASTQ files

• FASTQ files:
  1. human (files from the same group are merged together): /home/rothlab/rli/01_ngsdata/PPS_data/Human_pool/merged_pool9-1/
  2. yeast (files from the same plate are merged together): /home/rothlab/rli/01_ngsdata/PPS_data/yeast_pps_fastq/yeast_pps_fastq/

Install and Run

• install the package using: pip install PlasmidPoolAnalysis

  usage: pps [-h] [--align] [-f FASTQ] [-n NAME] -o OUTPUT -r REF
         [--refName REFNAME] [--summaryFile SUMMARYFILE] [--orfseq ORFSEQ]
  Plasmid pool sequencing analysis
  
  required arguments:
  -f FASTQ, --fastq FASTQ
                      path to fastq files
  -o OUTPUT, --output OUTPUT
                      Output directory
  -r REF, --ref REF     Path to reference
  -m MODE, --mode MODE  human or yeast
  --summaryFile SUMMARYFILE
                      Yeast or Human summary file
  
  optional arguments:
  -h, --help            show this help message and exit
  --align               provide this argument if users want to start with
                      alignment, otherwise the program assumes alignment was
                      done and will analyze the vcf files.
  -n NAME, --name NAME  Run name (default set to pps)
  
  --refName REFNAME     grch37, grch38, cds_seq. Required if mode == human
  -l LOG, --log LOG logging mode, default set to info
  

• Example: Human (with alignment to grch37)

  pps -f ~/01_ngsdata/PPS_data/Human_pool/merged_pool9-1/ -o ../../output/ -n Human91 --refName human91 --summaryFile ../../target_orfs/human_summary.csv -m human -r ../../fasta/human_91/ --align

• Yeast

  pps -f ~/01_ngsdata/PPS_data/yeast_pps_fastq/yeast_pps_fastq/ -o ../../output/ -n testpackYeast --summaryFile ../../target_orfs/yeast_summary.csv -m yeast -r ../../fasta/yeast_ref_all/

• The pipeline first submit alignment jobs to the cluster (slurm), after all the jobs are done, it filters vcf files, output summary and mutations

Output

• All the intermediate files will be saved into your output directory, a new folder will be made with the -n parameter
• For each fastq file, a folder will be made. It contains the following files:
  1. *.sh: alignment job script used for alignment
  2. all_summary_plateORFs.csv: summary for this plate/group
  3. *.log: log file
  4. *_raw.vcf: raw vcf file generated from pileup
  5. *_variants.vcf: vcf file with variants only
  6. *_filtered.vcf: filtered vcf file
• After the run is finished, the following files will be generated in the master output folder:
  1. alignment_log.csv: shows the alignment rate for each plate/group
  2. all_mutations.csv: contains all the variants passed filter
  3. all_summary.csv: contains all ORFs and if they were found/fully covered in the sequencing
  4. genes_stats.csv: overall stats