sedef_smk

Snakemake pipeline for running SEDEF on haplotype-resolved assemblies and generating final segmental duplication BED and BigBed files.

This workflow is adapted from DongAhn’s segmental duplication pipeline and is mainly intended for internal use in the lab.

---

Overview

This pipeline takes:

• haplotype FASTA files
• RepeatMasker BED
• TRF BED

and performs:

1. FASTA preparation and indexing
2. WindowMasker
3. repeat BED processing
4. FASTA masking
5. SEDEF
6. downstream filtering
7. final BED and BigBed generation

---

Input

The pipeline expects a tab-delimited manifest.tab file in the working directory.

Required columns:

• SAMPLE
• H1
• H2
• RM
• TRF

Example:

SAMPLE	H1	H2	RM	TRF
ASM  ASM/hap1.fasta	ASM/hap2.fasta	ASM/RM.bed.gz	ASM/trf.bed.gz

---

Output

Outputs are written under:

results/{sample}/

Main final outputs:

• results/{sample}/final_outputs/beds/{hap}.SDs.bed
• results/{sample}/final_outputs/beds/{hap}.SDs.lowid.bed
• results/{sample}/final_outputs/bigbeds/{hap}.SDs.bb
• results/{sample}/final_outputs/bigbeds/{hap}.SDs.lowid.bb

---

Run

Local

./runlocal 8

Cluster

./runcluster 30

---

Notes

• This pipeline is mainly intended for the Eichler lab compute environment.
• Contig names containing # are temporarily rewritten during intermediate steps and restored in final outputs.
• RM and TRF are expected to be RepeatMasker and TRF BED outputs could be generated by the Rhodonite annotation workflow https://github.com/vollgerlab/Rhodonite.
• Sample and haplotype expansion is driven by manifest.tab.

---

Repository structure

.
├── Snakefile
├── runlocal
├── runcluster
├── rules/
├── scripts/
├── schema/
└── resources/