To install dependencies, use the following command:
conda install -f addseq.yml
This repository contains scripts and data processing tools for analyzing chromatin accessibility using SMAdd-seq. Below is a breakdown of the directory structure:
./scripts/bash/- Bash scripts for data preprocessing, model training, and modification prediction using NEMO../scripts/- Jupyter notebooks for data processing and figure generation:- [0-7] - Notebooks for generating data output.
- [Figure1-3] - Notebooks for generating figures in the manuscript.
- [Supp_data] - Notebooks for generating supplemental data in the manuscript.
./figures/- Figures generated from the analysis.
- Basecalling Basecall raw sequencing data using Dorado:
~/dorado-0.7.3-linux-x64/bin/dorado basecaller \
/private/groups/brookslab/gabai/tools/[email protected] chrom.pod5 \
--emit-moves \
--device cuda:all \
--reference ./data/ref/sacCer3.fa > chrom.bam
- Quality Control (QC) Filter primary alignments, sort, and index BAM files using Samtools:
samtools view -b -F SECONDARY,SUPPLEMENTARY chrom.bam | samtools sort -o chrom_pass.sorted.bam
samtools index chrom_pass.sorted.bam
samtools stats chrom_pass.sorted.bam > chrom_pass.sorted.stats
- Event Alignment Align events to a reference genome using Uncalled4:
uncalled4 align \
--ref ./data/ref/yst/sacCer3.fa \
--reads chrom.pod5 \
--bam-in chrom_pass.sorted.bam \
-p 8 \
--eventalign-out chrom_pass.sorted.tsv \
--eventalign-flags print-read-names,signal-index,samples \
--pore-model dna_r9.4.1_400bps_6mer
- Ensure that Dorado and Samtools are properly installed before running the pipeline.
- Modify the paths to match your data storage and reference genome locations.
- The pipeline is optimized for Nanopore sequencing with R9.4.1 flow cells.