Table 1.
Features used in the ML classification algorithm
| Mutation probability | A count (low mapq) | A prior nucleotide | SNV base quality (high mapq) |
|---|---|---|---|
| Read depth (high mapq) | C count (low mapq) | T PRIOR NUCLEOTIDE | SNV base quality (high and low mapq) |
| Read depth (low mapq) | G count (low mapq) | C prior nucleotide | SNV mapping quality (high mapq) |
| Unmapped (forward) | T count (low mapq) | G prior nucleotide | SNV mapping quality (high and low mapq) |
| Unmapped (reverse) | A reference | A following nucleotide | SNV base quality read count (high mapq) |
| Soft-clipping read depth | T reference | T following nucleotide | SNV mapping quality read count (high mapq) |
| A count | C reference | C following nucleotide | SNV read count (high and low mapq) |
| C count | G reference | G following nucleotide | SNV position in read |
| G count | A SNV | A and soft-clipping | SNV forward strand |
| T count | T SNV | C and soft-clipping | |
| Repeat region | C SNV | G and soft-clipping | |
| Nearby SNV count | G SNV | T and soft-clipping |