RefGenPrelim

Documented code repository in response to Prelimenary Exam for Amelia Loeb Given by Dr. Amanda Hulse-Kemp May 2026

The goal: investigate differences of reference genomes across ploidy levels for sweetpotato The process:

• 1: download reference genomes reference genomes: I. trifida (NSP306) & I.batatas c.v. Beauregard v4 [I.batatas c.v. Taizong 6 included but not discussed]
• 2: record MD5SUM for reference genomes
• 3: Sample reads from reference genomes
  • simulated illumina reads
  • simulated PacBio wgs reads
• 4: Align reads back to their reference genomes

The scripts & code:

• 1.Download reference genomes
  • downloadrefgen.sh
  • Record md5 sum of utilized genomic references.
  • indexreference.sh

2. Simulate data of known genomic location for each reference genotype for the following:
   • a. Illumina short-read WGS data (to be equivalent to your DArT reads, 81bp I think)
     • simiss.sh or simiss.sh to simulate 200 million HiSeq reads per genome
       • 200 million was chosen to provide coverage for genome size
       • HiSeq simulation chosen based on https://pmc.ncbi.nlm.nih.gov/articles/PMC5407923/
       • did not work well because sampling location was not correctly saved in header
     • simwg.sh to simulate illumina 81bp reads, other lengths tested, including paired-ends
   • b. Pacbio long-read WGS data
     • simpb3.sh to simulate pacbio long reads. Did not work well because sampling location was unable to be identified
   • c. Optional – GBS and/or your actual DArT positond
     • aligned unique reference and alternate consensus sequences from population that was genotyped with DaRT panel
     • simgbs.sh
3. Align simulated data back to the following genome assemblies
   • a. Short Reads
     • iss_qcpre.sh optional quality control check
     • trim_iss.sh optional trimming of reads
     • aligniss.sh to align InSilicoSeq reads using bwa-me
     • tsviss.sh extract mapping coordinates from bam file
     • master_wgsim_pipeline.sh total script to align and extract mapping coordinates from simulated short reads produced by simwg.sh
   • b. long reads
     • alignpb.sh align pacbio reads simulated from simpb3
     • tsvpb.sh extract mapping coordinates from bam files of long reads
4. Visualize the proportion of reads aligning to the correct spot across the genomes
5. Additional Scripts [mostly unused]
   • mummer.sh tried aligning long reads to reference genomes, issue with awk commands for reads, did not use
   • qualimapat.sh tried to examine the quality of the mapping for short reads, did not prove insightful since coverage was not a factor
   • qcgenome.sh* script to compare the quality of reference genomes using assembly stats, busco & quash. Ran, but did not include data due to time

RefGenPrelim/
├── data/                       # Raw and processed genomic data
│   ├── bams/                   # Alignment files
│   ├── gbs_sim/                # GBS simulation outputs
│   ├── genomes/                # Reference FASTA files & MD5
│   ├── iss/                    # InSilicoSeq intermediate files
│   ├── pbsim3/                 # PacBio simulation data
│   ├── qualimap/               # Quality control reports
│   ├── stats/                  # Summary statistics
│   ├── trimmed_reads/          # Post-QC sequencing reads
│   └── wgsim/                  # wgsim simulation outputs
├── scripts/                    # Pipeline automation
│   ├── aligniss.sh
│   ├── alignpb.sh
│   ├── downloadrefgen.sh
│   ├── indexrefgenome.sh
│   ├── iss_qcpre.sh
│   ├── master_wgsim_pipeline.sh
│   ├── mummer.sh
│   ├── qcgenome.sh
│   ├── qualimapat.sh
│   ├── simgbs.sh
│   ├── simiss.sh
│   ├── simisspars.sh
│   ├── simpbs3.sh
│   ├── trim_iss.sh
│   ├── tsviss.sh
│   └── tsvpb.sh
├── results/                    # Final analysis outputs
├── logs/                       # Pipeline run logs
├── AssemblyStats.yml           # Assembly metrics config
├── README.md                   # Project documentation
└── RefGenPrelimEnv.yml         # Conda environment definition