Raw plot data for 'Nucleotide-by-nucleotide' plot:


Likelihood scores:

1. Sequence 1 ID number. (Always equal to '1' - i.e. the reference sequence.)
2. Sequence 2 ID number. (Order is the same as the rows in the 'Statistics summary' table, starting with '2'.)
3. Alignment coordinate (first nucleotide in the alignment is '1'.)
4. Log likelihood score for the null model.
5. Log likelihood score for the alternate model.

Note that entries for alignment coordinates corresponding to nucleotides which are gapped (or ambiguous nt codes) in either sequence 1 or sequence 2, or which are members of codons involved in stop to non-stop codon transitions, in either the null or alternate model CDSs, are omitted.


Likelihood scores summed over phylogenetic tree:

1. Alignment coordinate.
2. Number of contributing sequence pairs (e.g. not gapped at this position).
3. Sum of pairwise sequence divergences (mean number of mutations per nucleotide) for the contributing sequence pairs, divided by four.
4. Sum of log likelihood scores for the null model for the contributing sequence pairs, divided by four.
5. Sum of log likelihood scores for the alternate model for the contributing sequence pairs, divided by four.
6. Reference sequence coordinate.
7. Reference sequence nucleotide.

Note that, outside the query region, columns 2-5 are zero. See this note on summing scores over a phylogenetic tree; in particular for why columns 3-5 are divided by four.


Stop codon positions:

1. Model (0 = null model, 1 = alternate model).
2. Sequence ID.
3. Alignment coordinate.

Start codon positions:

1. Model (0 = null model, 1 = alternate model).
2. Sequence ID.
3. Alignment coordinate.

Gap (and ambiguous nt) positions:

1. Sequence ID.
2. Alignment coordinate.