Function to generate a homologous OmniPath resource
Usage
generate_homologs(
op_resource,
target_organism,
max_homologs = 5,
.missing_fun = NULL,
symbols_dict = NULL,
columns = c("source_genesymbol", "target_genesymbol"),
verbose = TRUE
)Arguments
- op_resource
a resource in the format of OmniPath/LIANA
- target_organism
`ncbi_taxid` or `name` of the target organism. See `show_homologene` for available organisms via OmnipathR's `HomoloGene`
- max_homologs
Determines the max number of homologs to be translated. Certain genes will have multiple homolog matches, with some having also certain isoforms considered. To exclude cases in which the number of matched homologs is too high, one can adjust the homologs parameter. Setting this to `1` would mean that one-to-many homolog matches are discarded
- .missing_fun
approach to handle missing interactions. By default set to `NULL` which would mean that any interactions without a homology match will be filtered. This can be set to e.g. `str_to_title` when working with murine symbols. Then if a gene has no matched homolog, instead of discarding it, the `.missing_fun` will be used to format the name from human. Hence, increasing the number of matches, but likely introducing some mismatches.
- symbols_dict
`NULL` by default, then `get_homologene_dict` is called to generate a dictionary from OmniPathR's homologene resource. Alternatively, one can pass their own symbols_dictionary.
- verbose
logical for verbosity
- source
name of the source (ligand) column
- target
name of the target (receptor) column