Parameters

There are a variety of parameters that you can alter by opening the script run_mlrgd in a text-editor and editing the Options section. A brief description of these are given below (suggested defaults are give by param = default).

1. pairs = 1: Method for generating the list of sequence pairs to use: 2 $\rightarrow$ use phylogenetic tree built with ednapars (maximum parsimony); 1 $\rightarrow$ use phylogenetic tree built with ednaml (maximum likelihood); 0 or other $\rightarrow$ use user-input file prefix.pairs.
2. refpos = 0: 0 $\rightarrow$ use seperate CDS files for each sequence; 1 or other $\rightarrow$ use reference sequence CDS files for all sequences.
3. gbkpos = 0: 0 $\rightarrow$ get CDS files from seqname.fasta.orfs (or seqname.gbk.orfs) files; 1 or other $\rightarrow$ get CDS files from seqname.gbk files (for any seqname.fasta files, the seqname.fasta.orfs file, if any, will be used).
4. skip = 0: 1 or other $\rightarrow$ skip sequence files containing non-ACGTUacgtu characters (exits if the reference sequence contains non-ACGTUacgtu characters); 0 $\rightarrow$ read standard ambiguous nt codes as N and ignore for most statistics (ambiguous nt are logged along with the gaps; codons containing ambiguous nt are logged along with the `zero-probability' transitions).
5. window1 = 5: $n$, where the running mean sliding window size for the 1st, 2nd and 3rd codon positions and for the 4-fold degenerate neutral sites plots is $2 n + 1$ nt (or codons).
6. window2 = 10: $n$, where the running mean sliding window size for the all sites and for the non-coding sites plots is $2 n + 1$ nt.
7. circular = 1: 0 $\rightarrow$ not a circular genome; 1 or other $\rightarrow$ circular genome.
8. clip1 = 0.00: Upper clipping threshold for clipped running mean: clip1 = 0.05 means that the upper 5% (i.e. columns with low observed number of mutations relative to expected number of mutations) in each window are clipped before calculating the mean.
9. clip2 = 0.30: Lower clipping threshold for clipped running mean: clip2 = 0.05 means that the lower 5% (i.e. the columns with high observed number of mutations relative to expected number of mutations) in each window are clipped before calculating the mean.
10. tttmin = 0.0: Lowest $t$ value to use in $t$-fitting.
11. tttmax = 10.0: Highest $t$ value to use in $t$-fitting.
12. tVfit = 0.2: Required fitting accuracy (total number of mutations between a sequence pair) for $t$ and $V$.
13. maxiter = 20: Maximum number of fitting iterations for $t$ and $V$.
14. Vmin = 0.01: Lowest $V$ value to use in $V$-fitting ($\geq$ 0.01).
15. Vmax = 10.0: Highest $V$ value to use in $V$-fitting ($\leq$ 10).
16. Vtype = 1: 0 $\rightarrow$ find seperate $V$ value for each sequence pair; 1 or other $\rightarrow$ find one $V$ value for the alignment.
17. Vfix = 0: If $>$ 0, use this $V$ value for all sequence pairs (defaults to 1 if $\leq$ 0; overridden if fitwhat = 3).
18. fitwhat = 3: 0 or other $\rightarrow$ fit $t$ using total number of mutations; 1 $\rightarrow$ fit $t$ using number of neutral/synonymous mutations; 2 $\rightarrow$ fit $t$ using number of neutral/synonymous mutations at 4-fold degenerate sites; 3 $\rightarrow$ as for 2, then adjust $V$ to fit the total number of mutations (i.e. including nonsynonymous mutations). Note: 1--3 are not recommended if the sequence is predominantly double-coding since, in this case, the number of neutral sites is potentially very small.
19. wholeseq = 1: Nucleotide range to use for all model-fitting, statistics and plots: 0 $\rightarrow$ use the region range1--range2, 1 or other $\rightarrow$ use the whole sequence.
20. range1 = 0: Start nucleotide of region to analyse (if wholeseq = 0; otherwise ignored), in reference sequence coordinates.
21. range2 = 0: End nucleotide of region to analyse (if wholeseq = 0; otherwise ignored), in reference sequence coordinates.