Double Strand Break Analysis Program
Analyze .fastq files for insertions, deletions, and microhomologous end joining at specified break site within reference sequence. Example output:
1. Install prerequisites
- Download and install Anaconda (https://www.anaconda.com/products/individual)
- Download and install git (https://git-scm.com/)
- Download and install samtools (http://www.htslib.org/download/)
2. Clone the git repo and cd into the directory (This step only needs to be done the first time.)
git clone https://github.com/lingchen42/dsbra.git
cd dsbra
3. Install dependencies within a conda virtual enviroment (This step only needs to be done the first time. If the dsbra virtual enviroment already exists, please remove the old one or give the new one a different name, using the -n flag.)
# Create a virtual environment named dsbra
conda env create -n dsbra python=3.7
# Activate the virtual enviroment**
conda activate dsbra
# Install r dependencies
conda install -c r-ggplot2 r-optparse
# Install python dependencies
pip install -r requirements.txt
5. Run the dsbra_analysis.py to infer the repair patterns
usage: dsbra_analysis.py [-h] -o OUT_DIR
[--align_mode {global,local}] --pcr_primer1
PCR_PRIMER1 --pcr_primer2 PCR_PRIMER2
--pcr_tail_seq PCR_TAIL_SEQ
[--min_len MIN_LEN] [--not_valid_reads]
[-r REF_FA] [-q FASTQ] [-b BREAK_INDEX]
[-m MARGIN] [-qm Q_REPAIR_MARGIN]
[--last_margin LAST_MARGIN]
[--count_cutoff COUNT_CUTOFF COUNT_CUTOFF]
[--full_table]
Double strand break repair analyzer (DSBRA)
optional arguments:
-h, --help show this help message and exit
-o OUT_DIR, --out_dir OUT_DIR
output directory
--align_mode {global,local}
choose the method for alignment
--pcr_primer1 PCR_PRIMER1
primer seq, at the begining of the ref seq; will
require the read to starts with this primer seq
--pcr_primer2 PCR_PRIMER2
primer seq, at the end of the ref seq; will require
last 4 nucleotide of the read is part of this primer
seq
--pcr_tail_seq PCR_TAIL_SEQ
added sequence for PCR;default
ATCGGAAGAGCACACGTCTGAACTCCAGTCAC
--min_len MIN_LEN miminal valid length (reads - pcr_tail_seq); default
44bp
--not_valid_reads output not valid reads
-r REF_FA, --ref_fa REF_FA
path of the reference sequence fasta file
-q FASTQ, --fastq FASTQ
path of the reads to analyze in fastq format
-b BREAK_INDEX, --break_index BREAK_INDEX
break index. 0-indexed, the position ofnucleotide
right before the break site.
-m MARGIN, --margin MARGIN
the margin before and after the break indexthat should
be considered for analysis;default 5bp.
-qm Q_REPAIR_MARGIN, --q_repair_margin Q_REPAIR_MARGIN
the margin before and after the last mutationthat
should be reportedin the repaired_read_sequence;
default 5bp.
--last_margin LAST_MARGIN
the margin outside of the last mutation eventthat
should be included in the repair sequencepattern for
those compound mutation events; default 5bp
--count_cutoff COUNT_CUTOFF COUNT_CUTOFF
take probability cutoff, number of colonies,apply
count cutoff to the summary tablebased on the poisson
distribution;default 0.001, 300
--full_table Will generate a large table withrepair pattern for
each analyzed
Example use:
ref_seq=NFR
break_idx=85
pcr_primer1=CGTAGGAAGTAGATTGTGTTAG
pcr_primer2=-1
pcr_tail_seq=ATCGGAAGAGCAACGTCTGAACTCCAGTCAC
margin=10
n_colony=300
poisson_cutoff=0.001
python dsbra_analysis.py -o rif1_nfr_wt/ -q example_data/RIF1_NFR.fastq -b $break_idx -m $margin --ref example_data/nfr_ref_seq.fa --pcr_primer1 $pcr_primer1 --pcr_primer2 $pcr_primer2 --pcr_tail_seq $pcr_tail_seq --not_valid_reads --full_table
The results will be in the rif1_nfr_wt directory as specified by -o, including:
RIF1_NFR_b84_m5_repair_pattern_summary_table.txtThis is the main result table, summarizing the dsbra repair pattern observed in the reads.RIF1_NFR_b84_m5_repair_pattern_full_table.txtThis has the repair pattern for each read. If we want to confirm how the read is converted to corresponding repair pattern, we can check this table.RIF1_NFR_b84_m5_run_info.txtThis is a text file documenting the setting of the script.RIF1_NFR.sam*These are the intermediate files generated by samtools when align the reads in the FASTQ file to the reference.RIF1_NFR_failed_alignment.samThis is the SAM file for failed alignments.RIF1_NFR_b84_m5_ref.faThis is the intermediate file generated by the script for storing the reference sequences.RIF1_NFR_b84_m5_mut_reads.sam*,RIF1_NFR_b84_m5_non_mut_reads.samThese are the intermediate filse generated by the script to store the reads with mutation within range of interest.
6. To plot the results, run the dsbra_grapher.py.
usage: dsbra_grapher.py [-h] [-R R] [-o OUT_DIR] -i INPUT_FILES
[INPUT_FILES ...] [--run_info RUN_INFO]
[--align_stats] [--all] [--mut_type]
[--del_len] [--del_seq] [--ins_len]
[--ins_seq] [--repair_seq]
[--aligned_mutations]
[--ref_bottom REF_BOTTOM] [--ref_top REF_TOP]
[--break_index BREAK_INDEX] [--fts FTS]
[--count_cutoff COUNT_CUTOFF COUNT_CUTOFF]
Visualize Mutation Events
optional arguments:
-h, --help show this help message and exit
-R R R plotting script path
-o OUT_DIR, --out_dir OUT_DIR
output directory, default is plots_2020-08-14/
-i INPUT_FILES [INPUT_FILES ...], --input_files INPUT_FILES [INPUT_FILES ...]
the output txt files from sam_dsbra_interface.py
--run_info RUN_INFO the run_info csv files from sam_dsbra_interface.py
--align_stats plot run_info.csv, plot alignment stats by Type
--all plot all types of plots
--mut_type plot Mutation Event Frequency by Type
--del_len plot Frequency of Deletions by Length
--del_seq plot Sequences with Deletion Event
--ins_len plot Frequency of Insertions by Length
--ins_seq plot Sequences with Insertion Event
--repair_seq plot top most common repair sequences
--aligned_mutations plot frequency and aligned mutations
--ref_bottom REF_BOTTOM
the start reference base to show aligned mutation
events
--ref_top REF_TOP the start reference base to show aligned mutation
events
--break_index BREAK_INDEX
the index of break site
--fts FTS axis font size; default 6
--count_cutoff COUNT_CUTOFF COUNT_CUTOFF
take probability cutoff, number of colonies,apply
count cutoff to the summary tablebased on the poisson
distribution;default 0.001, 300
Example use:
If we want to plot all types of plots:
python dsbra_grapher.py --all -i example_outputs/RIF1_NFR_output.txt --break_index ${break_idx} --count_cutoff $poisson_cutoff $n_colony -o example_outputs/ --run_info example_outputs/RIF1_NFR_first100k_global_aln_b${break_idx}_m${margin}_run_info.csv --fts 5
The plots will be in example_outputs directory.
More options can be found in ./sam_dsbra_grapher_ling.py -h, such as setting a count cutoff.
7. Finally, exit the virtual environment
conda deactivate
Soong, C. P., Breuer, G. A., Hannon, R. A., Kim, S. D., Salem, A. F., Wang, G., . . . Bindra, R. S. (2015). Development of a novel method to create double-strand break repair fingerprints using next-generation sequencing. DNA Repair (Amst), 26, 44-53. doi:10.1016/j.dnarep.2014.12.002