CiFi Computational Pipeline

1. Create nextflow Conda env and activate:

Conda will be used to install software. The pipeline relies on singularity and nextflow . Installation for both of these as well as any necessary modules are included in the conda environment.

wget https://raw.githubusercontent.com/mydennislab/CiFi/refs/heads/main/CiFi_Nextflow_env.yml
conda env create -f CiFi_Nextflow_env.yml
conda activate CiFi_Nextflow_env

2. Create paths to tmp files

By default the pipeline will store tmp files in /tmp. If you wish to write to new temporary files for SINGULARITY_TMPDIR and SINGULARITY_CACHEDIR save them here:

export SINGULARITY_TMPDIR=/PATH/TO/TMPDIR
export SINGULARITY_CACHEDIR=/PATH/TO/CACHE

3. Create path to results folder

mkdir /PATH/TO/RESULTS/FOLDER_NAME

When complete this folder will contain these folders:

• bams

• execution

• filtered_out

• ingress_results

• paired_end

As well as this report: wf-pore-c-report.html

4. Run nextflow command and check flags

The initial processing will use the wf-pore-c pipeline. See pipeline here \

nf-core framework :https://doi.org/10.1038/s41587-020-0439-x

Check flags:

nextflow run epi2me-labs/wf-pore-c \--help

Run Command:

nextflow run epi2me-labs/wf-pore-c -work-dir ./workflow/  -c  nextflow.config \
--bam <INPUT_BAM> \ 
--ref <PATH_TO_REF_GENOME> \
--cutter <RESTRICTION ENZYME NAME> \
--out_dir <./results/FOLDER_NAME> \
--threads 120 \
--minimap2_settings '-x map-pb' \
--paired_end \ #generates paired-end file that will be converted to Hi-C
--sample <SAMPLE_NAME> # can match folder

5. Filter the paired-end bam by mapping quality (MAPQ)

This software needs to be downloaded separated. See github here

Mapq filter only retains reads where BOTH reads in the pair cross the mapq threshold as well as removing unpaired reads

Download software

Environment Setup

mamba create -n bamfilter bioconda::samtools bioconda::htslib zlib gcc 

Compilation

g++ -O2 -o filter_bam bam_filter.cpp -I$CONDA_PREFIX/include -L$CONDA_PREFIX/lib -lhts -lz -lpthread -Wl,-rpath,$CONDA_PREFIX/lib

Usage

Run the program with the following command:

./filter_bam input.bam output.bam MAPQ_threshold [num_threads]

• input.bam: Path to the input BAM file
• output.bam: Path for the output BAM file
• MAPQ_threshold: Minimum MAPQ value to keep a read pair
• num_threads (optional): Number of threads to use (default is 1)

Run within the paired_end folder within results with a mapq filter of 1:

./filter_bam </RESULTS/FOLDER_NAME/paired_end/input.paired_end.bam> </RESULTS/FOLDER_NAME/paired_end/output.mapq_filtered_paired_end.bam> <1> <num_threads>

6. Convert filtered paired-end bam to pairs

Reactivate CiFi_Nextflow_env conda environment:

conda activate CiFi_Nextflow_env

Generate chromosome sizes file from index fasta reference

cut -f1,2 fasta.fai > sizes.genome

Run bam2pairs to convert filtered paired-end file into .pairs.gz file

bam2pairs -l -c <sizes.genome> <output.mapq_filtered_paired_end.bam> <sample.bam2pairs>

7. Convert pairs file to HiC file

Install java and load

Download juicertools

Juicertools takes in pairs input to output a .hic file. This .hic file can be loaded into Juicebox or UCSC for visualization.

Run juicertools:

java -jar juicer_tools.VERSION_jcuda.0.8.jar pre -t </PATH/TO/WORKING/DIR> <sample.bam2pairsbsorted.pairs.gz> <output_sample_name.hic> <sizes.genome>

8. Downstream analysis of CiFi data

Download CiFi_analysis.yml and activate:

wget https://raw.githubusercontent.com/mydennislab/CiFi/refs/heads/main/CiFi_analysis.yml
conda env create -f CiFi_analysis.yml
conda activate CiFi_analysis

Downloading CiFi Data

GM12878 CiFi HiFi v2.0 with DpnII as a restriction enzyme

GM12878_v2_path="https://bioshare.bioinformatics.ucdavis.edu/bioshare/download/rlc692m7tk5cibb/ciFi/GM12878_CiFi_v2/"

wget ${GM12878_v2_path}GM12878_DpnII_CiFi.HiFi_v2.0_hifi_reads.bam #Raw CiFi HiFi reads
wget ${GM12878_v2_path}GM12878_DpnII_CiFi.HiFi_v2.0_hifi_reads.bam.pbi #Raw CiFi HiFi reads index
wget ${GM12878_v2_path}GM12878_DpnII_CiFi.HiFi_v2.0_aligned_segments.cs.bam #Aligned (coordinate sorted) segments
wget ${GM12878_v2_path}GM12878_DpnII_CiFi.HiFi_v2.0_aligned_segments.cs.bam.csi #Aligned (coordinate sorted) segments index
wget ${GM12878_v2_path}GM12878_DpnII_CiFi.HiFi_v2.0_paired_end.ns.bam #Mock paired end bam

HG002 CiFi HiFi v2.0 with DpnII as a restriction enzyme

HG002_v2_path="https://bioshare.bioinformatics.ucdavis.edu/bioshare/download/rlc692m7tk5cibb/ciFi/HG002_CiFi_v2/"

wget ${HG002_v2_path}HG002_DpnII_CiFi.HiFi_v2.0_hifi_reads.bam #Raw CiFi HiFi reads
wget ${HG002_v2_path}HG002_DpnII_CiFi.HiFi_v2.0_hifi_reads.bam.pbi #Raw CiFi HiFi reads index
wget ${HG002_v2_path}HG002_DpnII_CiFi.HiFi_v2.0_aligned_segments.cs.bam #Aligned (coordinate sorted) segments
wget ${HG002_v2_path}HG002_DpnII_CiFi.HiFi_v2.0_aligned_segments.cs.bam.csi #Aligned (coordinate sorted) segments index
wget ${HG002_v2_path}HG002_DpnII_CiFi.HiFi_v2.0_paired_end.ns.bam #Mock paired end bam

GM12878 CiFi HiFi v3.0 with DpnII as a restriction enzyme

GM12878_v3_path="https://bioshare.bioinformatics.ucdavis.edu/bioshare/download/rlc692m7tk5cibb/ciFi/GM12878_CiFi_v3/"

wget ${GM12878_v3_path}GM12878_DpnII_CiFi.HiFi_v3.0_hifi_reads.bam #Raw CiFi HiFi reads
wget ${GM12878_v3_path}GM12878_DpnII_CiFi.HiFi_v3.0_hifi_reads.bam.pbi #Raw CiFi HiFi reads index
wget ${GM12878_v3_path}GM12878_DpnII_CiFi.HiFi_v3.0_aligned_segments.cs.bam #Aligned (coordinate sorted) segments
wget ${GM12878_v3_path}GM12878_DpnII_CiFi.HiFi_v3.0_aligned_segments.cs.bam.csi #Aligned (coordinate sorted) segments index
wget ${GM12878_v3_path}GM12878_DpnII_CiFi.HiFi_v3.0_paired_end.ns.bam #Mock paired end bam

GM12878 CiFi HiFi v3.0 with HindIII as a restriction enzyme

GM12878_v3_path="https://bioshare.bioinformatics.ucdavis.edu/bioshare/download/rlc692m7tk5cibb/ciFi/GM12878_CiFi_v3/"

wget ${GM12878_v3_path}GM12878_HindIII_CiFi.HiFi_v3.0_hifi_reads.bam #Raw CiFi HiFi reads
wget ${GM12878_v3_path}GM12878_HindIII_CiFi.HiFi_v3.0_hifi_reads.bam.pbi #Raw CiFi HiFi reads index
wget ${GM12878_v3_path}GM12878_HindIII_CiFi.HiFi_v3.0_aligned_segments.cs.bam #Aligned (coordinate sorted) segments
wget ${GM12878_v3_path}GM12878_HindIII_CiFi.HiFi_v3.0_aligned_segments.cs.bam.csi #Aligned (coordinate sorted) segments index
wget ${GM12878_v3_path}GM12878_HindIII_CiFi.HiFi_v3.0_paired_end.ns.bam #Mock paired end bam

HG002 CiFi HiFi v3.0 with DpnII as a restriction enzyme

HG002_v3_path="https://bioshare.bioinformatics.ucdavis.edu/bioshare/download/rlc692m7tk5cibb/ciFi/HG002_CiFi_v3/"

wget ${HG002_v3_path}HG002_DpnII_CiFi.HiFi_v3.0_hifi_reads.bam #Raw CiFi HiFi reads
wget ${HG002_v3_path}HG002_DpnII_CiFi.HiFi_v3.0_hifi_reads.bam.pbi #Raw CiFi HiFi reads index
wget ${HG002_v3_path}HG002_DpnII_CiFi.HiFi_v3.0_aligned_segments.cs.bam #Aligned (coordinate sorted) segments
wget ${HG002_v3_path}HG002_DpnII_CiFi.HiFi_v3.0_aligned_segments.cs.bam.csi #Aligned (coordinate sorted) segments index
wget ${HG002_v3_path}HG002_DpnII_CiFi.HiFi_v3.0_paired_end.ns.bam #Mock paired end bam

HG002 CiFi HiFi v3.0 with HindIII as a restriction enzyme

HG002_v3_path="https://bioshare.bioinformatics.ucdavis.edu/bioshare/download/rlc692m7tk5cibb/ciFi/HG002_CiFi_v3/"

wget ${HG002_v3_path}HG002_HindIII_CiFi.HiFi_v3.0_hifi_reads.bam #Raw CiFi HiFi reads
wget ${HG002_v3_path}HG002_HindIII_CiFi.HiFi_v3.0_hifi_reads.bam.pbi #Raw CiFi HiFi reads index
wget ${HG002_v3_path}HG002_HindIII_CiFi.HiFi_v3.0_aligned_segments.cs.bam #Aligned (coordinate sorted) segments
wget ${HG002_v3_path}HG002_HindIII_CiFi.HiFi_v3.0_aligned_segments.cs.bam.csi #Aligned (coordinate sorted) segments index
wget ${HG002_v3_path}HG002_HindIII_CiFi.HiFi_v3.0_paired_end.ns.bam #Mock paired end bam

GM12878 data is included in this study and to please cite upon use: https://www.biorxiv.org/content/10.1101/2025.01.31.635566v1.article-info

HG002 data remains unpublished. If planning to use in publication please reach out to the Dennis lab