- Python 100%
| Filename | Latest commit message | Latest commit date |
|---|---|---|
| src | ||
| README.md | ||
| requirements.txt | ||
PRIME-In_for_T_cell_engineering
This code implements our data preprocessing and analysis pipeline for PRIME-In_for_T_cell_engineering data. It takes a parameter manifest file (.yaml) with raw sequencing reads (FASTQ) info and produces a table of annotated gene knock-in types as output.
Dependencies
To ensure successful execution of the pipeline, the following dependencies are required:
- Python 3 — Programming language for executing the pipeline scripts
- Reference genome FASTA file (Example) — Used for alignment and annotation
FLASH—Tool for merging Paired-ends readsbwa— Alignment tool for mapping sequencing reads to the reference genometrimmomatic— Flexible read trimming tool for Illumina NGS datasamtools— Tool for manipulating alignments in the SAM/BAM formatblastn— Nucleotide-Nucleotide BLAST
Note on Path Configuration
If you see a warning when running the script saying that certain software cannot be found, it's likely due to the use of relative paths in the src directory. To fix this, you can manually replace the relative paths in the script with absolute paths specific to your environment.
Getting Set Up
Installation It's recommended (but not essential) to set up a conda environment to manage dependencies
conda create -n PEM python=3.8
conda activate PEM
git clone --recursive https://github.com/GUOSHU-COOL/PRIME-In_for_T_cell_engineering
cd PRIME-In_for_T_cell_engineering
pip install -r requirements.txt
Remove redundant sequences to avoid multiple alignments to the transgene containing human-derived fragments
cd src
wget http://hgdownload.soe.ucsc.edu/goldenPath/hg38/bigZips/hg38.fa.gz
gunzip hg38.fa.gz
python remove_specfic_seq_infa.py
Create each sample blastn db
To generate the BLASTN database, run:
python generate_index_add_transgene_fa.py
python generate_index_add_transgene_KI_fa.py
Running the Knock-in Type Classifier
To run the full knock-in type classification pipeline, you must first create a manifest YAML file that describes all pipeline inputs, including a data_info.xlsx file which contains sample-specific metadata.
Once your files are ready, simply run:
python batch_reads_type_classifier.py --config ./config.yaml
An example config.yaml and data_info.xlsx can be found in the src directory.
Required Columns in data_info.xlsx
The data_info.xlsx file should contain the following columns for each sample:
| Column Name | Description |
|---|---|
sample |
Sample name / identifier |
chr_seq |
Chromosome or sequence name to align against (e.g., chr1, chr7) |
cut_site |
Genomic coordinate (1-based) of the CRISPR cut site |
strand |
Strand direction: + or - |
primer_plus_5bp |
Primer sequence including 5 bp upstream flanking sequence |
primer_to_cut_plus20bp |
Subsequence from primer to cut site, plus 20 bp downstream from the cut |
Make sure all values are correctly formatted for each sample; otherwise, the pipeline may raise input parsing errors.
After the Knock-in Type classify, we analysis sgRNA-caused Off-target type
Running the sgRNA-caused Off-target analysis
python sgRNA_analysis.py