From an R terminal, domainArch can be installed using devtools:

if (!requireNamespace("devtools"))
    install.packages("devtools")
devtools::install_github("trvinh/domainArch", INSTALL_opts = c('--no-lock'), dependencies = TRUE)

Or you can download the source code and install it with the command:

install.packages(path_to_domainArch, repos = NULL, type="source")

To run domainArch, enter:

library(domainArch)
runDomainArch()

Currently the tool accepts 3 kinds of input:

• A folder containing several domain files, each file for a single seed protein together with its orthologs. These files are the *.domains file generated by fdog.run. Please note: each domain file must be named following this format <seed/groupID>.domains
• A single concatenated domain file for multiple seeds and ortholog proteins. This file is the output of fdogs.run or generated by combining multiple single fDOG outputs using fdog.mergeOutput function
• A folder containing annotation files for several species in JSON format, e.g. annotation_dir in fDOG data. These annotation files can be created using fas.doAnno function of FAS tool

In the future one can use directly the protein sequences as input. Stay tuned! ;)

Tip: if your concatenated domain file is too large, you should split it into multiple singe domain files by using the function Split domain file provided by domainArch.

For any bug reports or questions, please open an issue on GitHub or be in touch via email tran@bio.uni-frankfurt.de