Pangenome of cultivated beet and crop wild relatives

This repository contains scripts relevant for the investigation of the beet pangenome for:

A) employing k-mer- and gene-based methods to resolve the parental relationships of the tetraploid wild beet B. corolliflora

and

B) identifying regions not present in the cultivated sugar beet, but associated with traits with relevance for breeding.

---

A)

Different approaches were developed to investigate the parental relationships of the tetraploid wild beet B. corolliflora. The following scripts are provided in this repository:

• trio binning java application: /polyploid_parental_relations_analyses/SBTTrio/out/artifacts/trio_binning/SBTTrio.jar

• k-mer fingerprinting java application: /polyploid_parental_relations_analyses/SBTTrio/out/artifacts/fingerprinting/SBTTrio.jar

• generate artificial reads: /polyploid_parental_relations_analyses/generate_artificial_reads.py

• artificial read mapping: /polyploid_parental_relations_analyses/map_artificial_reads_distributed.py

B)

The script 'find_regions_of_interest.py' allows the identification of zero coverage regions as well as the identification of conserved regions in a species of interest.

As input, a tab separated file with chromosome/contig name (first column), position (second column), coverage at this position (third column) and variant (1) vs. no variant (0) at this position (fourth column) is required.

(Example line: chr1 1 0 0)

The coverage and variance thresholds allow the identification of regions with properties of interest. A 'max_cov' threshold of 0.01 means that the maximal allowed coverage in a window is 1% of the mean coverage over the whole chromosome/contig in a long read-based assembly. For larger genomes > 1 Gbp it is recommended to split the input file into two smaller files.

Example command to identify zero coverage regions:

python3 identify_lost_and_conserved_regions/find_regions_of_interest.py  --output_dir /out/name/ --cov_var /in/coverage_and_variants.txt --mode lost --assembly_type long --max_cov 0.01 --var 0 --window_size 2500 --shift 150

The identified regions (extended_regions.txt) can then be investigated for overlapping genes.