Input formats and shared conventions

The canonical command page is authoritative when a command has stricter requirements than the shared conventions below.

Multiple sequence alignments

PhyKIT detects alignment format from file content. Supported formats are FASTA, Clustal, MAF, Mauve, PHYLIP (interleaved, sequential, and relaxed), and Stockholm. Aligned sequences must have equal lengths.

FASTA records use the first whitespace-delimited token after > as the taxon identifier:

>species_A optional description
ACGTACGT
>species_B
ACGT-CGT

Use unique identifiers. Gap characters are normally -; the handling of ambiguous characters depends on the statistic being calculated.

Phylogenetic trees

Tree inputs use Newick unless a command page states otherwise:

((species_A:0.1,species_B:0.1):0.2,species_C:0.3);

Terminate each tree with a semicolon. Branch lengths and internal support values are required only by commands that use them. Some multi-tree commands accept either one Newick tree per line or one tree-file path per line.

Taxon names

Taxon matching is exact and case-sensitive. species_A, Species_A, and species A are different identifiers. Use the same spelling in alignments, tree tips, trait tables, grouping files, and mapping files. Avoid whitespace in identifiers when possible.

Single-trait tables

Commands such as phylogenetic_signal accept two tab-separated columns with no header. Blank lines and lines beginning with # are ignored:

# taxon<TAB>continuous_value
species_A    1.25
species_B    0.80

Multi-trait tables

Regression, ordination, and other multi-trait commands use a header. The first column contains taxa and subsequent columns contain named traits:

taxon        body_mass       brain_size
species_A    1.25    0.90
species_B    0.80    0.72

Use literal tab characters, not aligned runs of spaces. Numeric analyses reject non-numeric trait values unless the command explicitly supports categorical data.

Discrete traits and grouping files

Discrete-trait commands commonly use a header so --trait can identify the selected column:

taxon        diet
species_A    carnivore
species_B    herbivore

Other two-column files, such as rename maps, usually contain an existing name and its replacement. Consult the command page for column order and header rules.

Missing values and missing taxa

PhyKIT does not define one missing-value token for every command. Do not assume that NA, NaN, an empty field, or ? is accepted. Check the command's input section and runtime interface. Missing taxa are also command-specific: some commands use shared taxa, some expose --missing-taxa, and others reject nonmatching sets.

Paths and list files

Paths passed directly on the command line are resolved from the current working directory. For files containing lists of other file paths, consult the command page to determine whether relative paths are resolved from the working directory or the list file's directory. Absolute paths remove that ambiguity.