The dcGO database offers a comprehensive mapping of
protein domains
toontologies
.
The dcGO update in 2023 extends annotations for protein domains from different definitions/levels (SCOP, Pfam, and InterPro) using commonly-used ontologies (categorised into functions, phenotypes, diseases, drugs, pathways, regulators, and hallmarks). This update adds new dimensions to the utility of both ontology and protein domain resources.
A new website has been launched to provide users with a more integrated and user-friendly experience. This website features: enhanced faceted search capabilities that return term- and domain-specific information pages; improved ontology hierarchy capabilities that browse both ontology terms and their annotated domains; and a newly added facility that supports domain-based ontology enrichment analysis.
To learn how to use the dcGO database, a user manual has been made available here with step-by-step instructions.
The dcGO website was developed using a Perl real-time web framework Mojolicious.
The dcGO website was also developed using Bootstrap that supports a mobile-first and responsive webserver across all major platform browsers.
The folder
my_dcgo
has a tree-like directory structure with three levels:my_dcgo ├── lib │ └── My_dcgo │ └── Controller ├── public │ ├── app │ │ ├── ajex │ │ ├── css │ │ ├── examples │ │ ├── img │ │ └── js │ ├── dcGObooklet │ │ ├── index_files │ │ └── libs │ └── dep │ ├── HighCharts │ ├── Select2 │ ├── bootstrap │ ├── bootstrapselect │ ├── bootstraptoggle │ ├── dataTables │ ├── fontawesome │ ├── jcloud │ ├── jqcloud │ ├── jquery │ ├── tabber │ └── typeahead ├── script ├── t └── templates └── layouts
Assume you have a ROOT (sudo)
privilege on Ubuntu
sudo su
# here enter your password
curl -L cpanmin.us | perl - Mojolicious
perl -e "use Mojolicious::Plugin::PODRenderer"
perl -MCPAN -e "install Mojolicious::Plugin::PODRenderer"
perl -MCPAN -e "install DBI"
perl -MCPAN -e "install Mojo::DOM"
perl -MCPAN -e "install Mojo::Base"
perl -MCPAN -e "install LWP::Simple"
perl -MCPAN -e "install JSON::Parse"
perl -MCPAN -e "install local::lib"
perl -MCPAN -Mlocal::lib -e "install JSON::Parse"
sudo su
# here enter your password
# install R
wget http://www.stats.bris.ac.uk/R/src/base/R-4/R-4.2.0.tar.gz
tar xvfz R-4.2.0.tar.gz
cd ~/R-4.2.0
./configure
make
make check
make install
R # start R
sudo su
# here enter your password
# install pandoc
wget https://github.com/jgm/pandoc/releases/download/2.18/pandoc-2.18-linux-amd64.tar.gz
tar xvzf pandoc-2.18-linux-amd64.tar.gz --strip-components 1 -C /usr/local/
# use pandoc to render R markdown
R
rmarkdown::pandoc_available()
Sys.setenv(RSTUDIO_PANDOC="/usr/local/bin")
rmarkdown::render(YOUR_RMD_FILE, bookdown::html_document2(number_sections=F, theme="readable", hightlight="default"))
Assume you place my_dcgo
under your home
directory
cd ~/my_dcgo
systemctl restart apache2.service
morbo -l 'http://*:3080/' script/my_dcgo
For any bug reports, please drop email.