Phased Genome Assembly and Evaluation Pipeline

This repository contains a series of PBS scripts designed to perform a complete, phased genome assembly from PacBio HiFi and Hi-C reads. The pipeline uses hifiasm for assembly and includes steps for data preparation, quality control with BUSCO and Merqury, and contact map generation with pairtools and Pretext.

The scripts are structured to be modular and easily reusable for new projects.

---

Workflow Overview

The pipeline is broken down into sequential steps, with each step residing in its own numbered directory. Scripts should be executed in numerical order.

• 01_hifiBamToFasta: Converts the input PacBio HiFi BAM file into a gzipped FASTA format required by the assembler.
• 02_03_meryl_genomescope: Contains two scripts:
  • 02_merylRawHifi.pbs: Counts k-mers from the HiFi reads using Meryl.
  • 03_genomescope.pbs: Analyzes the k-mer spectrum with GenomeScope2 to estimate genome characteristics like size, heterozygosity, and ploidy.
• 04_hifiasm: Performs the main genome assembly using HiFi reads and scaffolds the contigs into chromosomes using Hi-C data.
• 05_gfastats: Converts the GFA (Graphical Fragment Assembly) output from hifiasm into standard FASTA files for the primary assembly and both haplotypes.
• 06_buscoContigs: Runs BUSCO (Benchmarking Universal Single-Copy Orthologs) to assess the completeness of the assemblies.
• 07_merquryContigs: Runs Merqury to evaluate assembly consensus quality (QV), completeness, and phasing based on k-mer content.
• 08_contactMatrixContigs: Aligns the Hi-C reads back to the final assemblies and generates contact maps to visually inspect the chromosome-scale scaffolding.

---

How to Use This Pipeline

This pipeline is designed to be easily run on a new dataset with minimal changes.

1. Clone the Repository

Clone this repository to your HPC environment.

git clone <your_repository_url>
cd <repository_name>

2. Add Your Raw Data

Place your raw sequencing files into the rawdata/ directory. You will need:

• PacBio HiFi reads (in BAM format)
• Paired-end Hi-C reads (e.g., *_R1.fastq.gz and *_R2.fastq.gz)

3. Configure and Run

For each script (01 through 08), you only need to edit the CONFIGURATION block at the top to match your new project.

Example: Editing 04_hifiasm.pbs for a new project named "Silver_Queen"

# ====================================================================
# --- CONFIGURATION: EDIT THIS VARIABLE FOR A NEW PROJECT ---
# ====================================================================

# 1. Define the unique prefix for your current project.
PROJECT_NAME="Silver_Queen"

# ====================================================================

# You would also update the paths to the input HiFi and Hi-C reads
# in the "GLOBAL VARIABLES" section of the script if they are different.

After configuring a script, submit it to the scheduler:

qsub 01_hifiBamToFasta/01_hifiBamToFasta.pbs

Wait for the job to complete successfully, then configure and submit the next script in the sequence.

---

Dependencies & Environment

This workflow was designed for a PBS-based HPC cluster and assumes the following:

• Conda: A working Conda installation (miniconda3 or anaconda3) is required to create the software environments.
• Software Modules: The pipeline uses custom-built software modules for hifiasm, gfastats, and meryl. Instructions for creating these modules are commented in the header of each respective script.
• Conda Environments: Several scripts require specific conda environments. The one-time setup commands are included in the header of each script (e.g., conda create -n busco_env -c bioconda busco).