VCF_annotation

Part I: Import database, libraries and functions

Part II: VCF file cleaning

Expand INFO and FORMAT to columns (Output: "vcf_extend.csv")

Part III: VCF file annotation

0. Initialize annotated dataframe: vcf_annotation

1. Type of variation and their effect

Score severity of variation effect
High risk (3): Nonsense mutation (Stop codon gain), Stop codon lost, Start codon lost, Frameshift Moderate risk (2): Missense mutation, Inframe insertion, Inframe deletion Low risk (1): Intergenic/Intragenic variant No risk (0): Silent mutation

Annotate variation effect

• Type: snp
  • If variation is in cds region:
    • NO: Intergenic/Intragenic variant
    • YES -> find codon -> if encoding amino acid is changed:
      • NO: Silent mutation
      • YES -> if this is start codon or stop codon:
        • NO: Missense mutation
        • YES: Nonsense mutation or Stop codon lost or Start codon lost
• Type: insertion or deletion
  • If variation is in cds region:
    • NO: Intergenic/Intragenic variant
    • YES -> if the insertion/deletion is multiple of 3:
      • NO: Frameshift
      • YES: Inframe insertion or Inframe deletion
• Type: mnp (length = 2)
  • If variation is in cds region:
    • NO: Intergenic/Intragenic variant
    • YES -> if two nucleotides are in the same codon:
      • YES: check codon as defined for snp
      • NO: check two codons individually as defined for snp, then select the most deleterious effect
• Type: complex
  • If variation is in cds region:
    • NO: Intergenic/Intragenic variant
    • YES -> if the length is changed in variant:
      • YES: follow steps defined for insertion/deletion
      • NO: check all related codons, then select the most deleterious effect

2. Depth of sequence coverage at the site of variation

3. Number of reads supporting the variant

4. Percentage of reads supporting the variant versus those supporting reference reads

5. Allele frequency of variant from ExAC API

Part IV: save annotated VCF files

Output: vcf_annotation.csv