MLCA Genotyping

This page details the methodology applied to define genogroups and genotypes within BTV-GLUE, a GLUE project focused on the comparative genomics of bluetongue virus (BTV).

1. Data Collection and Initial Sequence Processing

---

• Sequence Collation: All BTV sequences were collated from GenBank as of the date 30-06-2018.

• Exclusion of Sequences: Certain sequences were manually excluded from the analysis based on specific criteria. The accession numbers of these sequences and the reasons for their exclusion are documented in the following file:

  • BTV-GLUE/tabular/formatted/sequences_to_exclude.txt

2. Segment Assignment and Reference Selection

---

• Segment Assignment Procedure:

  • A combination of GenBank annotations and the BLAST-based btvSegmentRecogniser module was used to assign each sequence to its corresponding segment.

• Master Reference Sequences: The master reference sequences selected for each segment are as follows:

  • Segment 1: JX680457
  • Segment 2: JX680458
  • Segment 3: JX680459
  • Segment 4: JX680460
  • Segment 5: JX680461
  • Segment 6: JX680462
  • Segment 7: JX680463
  • Segment 8: JX680464
  • Segment 9: JX680465
  • Segment 10: JX680466

• Outgroup Sequences:

  • Epizootic Hemorrhagic Disease Virus (EHDV): AM744977-AM744986
  • Palyam Virus (PATAV): JQ070386-JQ070395

3. Sequence Length Criteria for Segment Inclusion

---

• Segments 1-9: Sequences must have a length of at least 90% of the segment master reference length to ensure the coding region is adequately captured.

• Segment 10: Due to the smaller proportion of the coding region, sequences need to be at least 80% of the segment master reference length.

4. Alignment Construction

---

• Nucleotide Alignment:

  • A nucleotide alignment (BTV_COMPL_SEG_NT) was constructed using MAFFT v7.299b (default settings) for each segment, incorporating the master reference sequence along with all non-excluded GenBank sequences.

• Protein Translation and Outgroup Integration:

  • Each sequence in the nucleotide alignment was translated into an unaligned protein sequence of its coding region.
  • Protein translations of the relevant outgroup sequences were combined and aligned using MAFFT v7.299b.

• Reverse Translation to Nucleotide Alignment:

  • The aligned protein sequences were reverse-translated back to nucleotide sequences using TBLASTN in GLUE's blastFastaAlignmentImporter pattern.
  • This produced BTV_OUTG_CODON alignments (outgroup codon alignments) that were inspected and manually curated.

5. Phylogenetic Tree Construction

---

• Building Phylogenetic Trees:

  • Phylogenetic trees were generated from these nucleotide alignments using RAxML-NG v0.8.1 with the GTR+G+I model and 500 bootstrap replicates.
  • The Transfer Bootstrap Estimate (TBE) method was employed for node support calculation (refer to https://doi.org/10.1038/s41586-018-0043-0).

• Tree Rerooting and Outgroup Usage:

  • Outgroups were generally used for tree rerooting, after which they were removed, except for segment 6, where the EHDV sequence was retained.
  • For segments 5, 8, 9, and 10, midpoint rooting was applied due to the unrealistic placement of outgroup rooting.

6. Genogroup and Genotype Demarcation

---

• ClusterPicker Settings:

  • ClusterPicker 1.2.5 was used to define genogroups and genotypes based on the following parameters:
    • Transfer Bootstrap Threshold: 75%
    • Genetic Distance Strategy: "gap"

• Genetic Distance Thresholds:

  • Segment 1 Genogroup: p = 0.2
  • Segment 2 Genogroup: p = 0.35
  • Segment 2 Genotype: p = 0.22
  • Segment 3 Genogroup: p = 0.2
  • Segment 4 Genogroup: p = 0.2
  • Segment 5 Genogroup: p = 0.2
  • Segment 6 Genogroup: p = 0.33
  • Segment 7 Genogroup: p = 0.2
  • Segment 8 Genogroup: p = 0.2
  • Segment 9 Genogroup: p = 0.2
  • Segment 10 Genogroup: p = 0.2

• These thresholds were chosen to reflect groupings from the literature, ensuring that previously characterized serotypes align with one or two genotypes for segment 2.

7. Reference Sequence Selection

---

• Manual Selection of Reference Sequences:
  • References were manually chosen by Kiki to capture the genetic diversity within each genogroup and genotype.
• Reference Sequence Files:
  • Detailed lists of reference sequences for each segment can be found in:
    • BTV-GLUE/tabular/formatted/Segment<X>RefList.txt (where <X> is the segment number).

8. Segment Reference Phylogenies

---

• Genotyping Codon Alignments:

  • Genotyping codon alignments were created by retaining only the reference sequences from the outgroup codon alignments for each segment.

• Tree Generation for Segment References:

  • RAxML 8.2.8 was used with the GTRGAMMAI model and 1000 bootstrap replicates to generate the segment reference phylogenies.

• Rooting of Segment Reference Phylogenies:

  • The trees were rooted at internal nodes that separate specific genogroups, for example:
    • Segment 1: Rooted at ["C", "D", "E"]
    • Segment 2: Rooted at ["G", "J"]
    • Segment 3: Rooted at ["C", "D", "E"]
    • (Additional details for each segment as provided in the original text)

9. Automated Typing Process

---

• Automated Genotype Assignment:
  • The genotyping codon alignments and segment reference phylogenies are the basis of the automated typing process available via the BTV-GLUE website.
• Method Employed:
  • This process uses the Maximum Likelihood Clade Assignment method described in the GLUE Bioinformatics paper.
• Configuration Parameters:
  • Distance Scaling Exponent: -3.0
  • Distance Cutoff: Equal to the p-distance threshold set earlier for each segment.
  • Internal Distance Cutoff: Twice the distance cutoff value.
  • Clade Category Cutoff: 80%

---