CoLoRGen: comprehensive long read genotyping pipeline. This tool generates consensus sequence of the targeted regio per allele and returns vcf files for each gene that is found. If star-alleles are present, the correct star-allles are assigned to the genes.
CoLoRGen can be installed by downloading these git repository and run the install script.
Using this method requires the user to provide several binaries:
samtools/bgzip/tabix version 1.11 and minimap2 version 2.18 are recommended
as these are those used in development of CoLoRGen.
Most of the script use python3, therefore python3 need to be installed on the system. Also the virtualenv package need to be present as CoLoRGen runs everything in a virtual environment.
Running CoLoRGen is easy. With the command below the full pipeline is run.
CoLoRGen -p <parameters.sh>
The following paramters need to be adjust in the parameters.sh file:
number_treats=40
nanopore_sequencing_files='fastq_pass' # raw fastq files
analyse_folder='./' # working directory
ref_folder="ref/" # reference folder
reference_genome='GRCh38.fa' # fasta file of the reference genome
minimap2_index='GRCh38.fa.mmi' # minimap2 index file
reference_gtf='GRCh38.gtf' # gtf file with the exon regions
# edit paths to external software tools
export PATH=samtools-1.11:$PATH
export PATH=bcftools-1.11:$PATH
export PATH=htslib-1.11:$PATH
hap_py='hap.py' #path to the hap.py script
gene_json='cyp2d6_hybrids.json' # json file with gene info and hybrid info
star_alleles="${ref_folder}CYP2D6.NC_000022.11.haplotypes.tsv" # star allele nomenclature downloaded from PharmVar
# position of the targeted region
target_start=42121165
target_stop=42149371
target_chr='chr22'
# medaka model
model_m='r941_min_hac_g507' # model for medaka variant calling and consensus
log_level='debug' # log level
In the git repository, there is a example of the json file with the gene info
The output is structured in different folder. All the intermediate files are kept for debugging en visualisation proposes.
.
├── ref # folder with genome and star-allele references
├── haplotype
│ ├── consensus_1 # first consensus based on the structural variants
│ ├── consensus_1_breakpoints.txt # break points defining the structural variants
│ ├── consensus_2 # second consensus constructed with the reads that mapped on the alleles defined by medaka
│ ├── consensus_2_H1 # consensus 2 folder for H1
│ ├── consensus_2_H2 # consensus 2 folder for H2
│ ├── consensus_3 # third consensus constructed with all the reads
│ ├── consensus_3_genes # target genes mapped on consensus sequence
│ ├── consensus_3_H1 # consensus 3 folder for H1
│ ├── consensus_3_H2 # consensus 3 folder for H2
│ ├── coverage.txt # depth on the target sequence
│ ├── haplotypes_H1_gene_*_*.vcf # vcf file of the genes on allele 1
│ ├── haplotypes_H2_gene_*_*.vcf # vcf file of the genes on allele 2
│ ├── haplotypes_haplotypes.txt # overview of the found star-alleles
│ ├── haplotypes.vcf # vcf file of all the found variants
│ ├── medaka_round_0 # medaka output of the input data
│ ├── select_reads_taged # reads split by haplotypes defined by medaka
├── log
│ ├── error.log # error log
│ └── stout.log # standard log
├── mapped # mapped data
└── raw_reads # raw reads
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent