Files

01_download_ISBT_Table.py

• Downloads ISBT Blood Alleles Tables using URLS provided in ISBT_links.txt.

02_extract_ISBT_table.py

• Extracts and cleans the Phenotype, Allele Name and Nucleotide Change columns in ISBT Blood Alleles Tables.
• Uses HGVS_Notation.txt to create Blood Group, Gene and Chromosome columns.
• Appends HGVS transcript notation in front of Nucleotide Change column.

03_ABO_add.py

• Adds missing nucleotide change to B alleles in the ABO blood group.

04_Erythrogene_table.py

• Generates a table with the same structure as the output of 02_extract_ISBT_table.py using erythrogene_alleles.xlsx.

05_get_coords.py

• Uses VariantValidator to:
  • Map the Nucleotide Change column in extracted ISBT and Erythrogene tables to GRCh37 and GRCh38 coordinates and VCF position.
  • Create two new columns GRCh37 Alt Allele and GRCh38 Alt Allele to represent the genotype value of the nucleotide change.

06_edit_genotype_value.py

• Adjust VCF positions for ABO:c.261G.
• Corrects genotype values for biallelic/multiallelic sites.

07_blood_allele_table.py

• Produces two final output files:
  • Blood_Allele_Table.tsv
  • Blood_Allele_Table_Separated.tsv (Exploded version)
• Uses the following files:
  • ISBT_variants_to_remove.tsv and Erythrogene_variants_to_remove.tsv to remove alleles with nucleotide change that cannot be found or automapped by VariantValidator, have variant reference not aligning with reference sequence or involves Exon/Intron deletion.
  • ISBT_variants_to_overwrite.tsv provides the mapping details of nucleotide change that cannot be found using 05_get_coords.py but works when using VariantValidator Web Interface

08_Infer.R

• Uses genotype data stored in chr[]_df_genotype.Rdata and allele information from Blood_Allele_Table_Separated.tsv to infer potential phenotypes at each variant position.

09_summary_table.R

• Condenses the results from 08_Infer.R into a summary table with blood group as columns and samples as rows.
• List the potential phenotypes for each blood group and sample.

10_final_summary_table.R

• From 09_summary_table.R, the phenotype with the most number of Nucleotide Change associate is inferred as the "correct" phenotype for the sample.