PhyloProfile, (R)Tool for phylogenetic profiling

!language: R]( !presented at: BOSC2017]( !GitHub release]( !license: MIT](

PhyloProfile is a Shiny-based tool for integrating, visualizing and exploring multi-layered phylogenetic profiles.

Alongside the presence/absence patterns of orthologs across large taxon collections, PhyloProfile allows the integration of any two additional information layers. These complementary data, like sequence similarity between orthologs, similarities in their domain architecture, or differences in functional annotations enable a more informed interpretation of phylogenetic profiles.

By utilizing the NCBI taxonomy, PhyloProfile can dynamically collapse taxa into higher systematic groups. This enables rapidly changing the resolution from the comparative analyses of proteins in individual species to that of entire kingdoms or even domains without changes to the input data.

PhyloProfile furthermore allows for a dynamic filtering of profiles – taking the taxonomic distribution and the additional information layers into account. This, along with functions to estimate the age of genes and core gene sets facilitates the exploration and analysis of large phylogenetic profiles.

Explore the installation-free online version or watch the demo video to learn more.

Table of Contents

Installation & Usage

PhyloProfile is based on the R-package Shiny, as such a recent version of R is needed. Once that is out of the way you can just clone this repository to get a copy of PhyloProfile.

git clone

To start PhyloProfile simply move into the PhyloProfile directory and run the main script

cd PhyloProfile
Rscript phyloprofile.R

The initial start can take a while, as phyloprofile.R will try do download and install all necessary dependencies automatically. (Note: Depending on your system this sometimes fails, please check the console log for error messages concerning the dependency installation)

Once all packages are downloaded and installed your web browser will open a new tab and display the main PhyloProfile menu.

Input Data

PhyloProfile can read a number of different input files, including regular tab-separated files and OrthoXML. The additional information layers can be embedded in the OrthoXML or be provided separately.


PhyloProfile is capable of reading in files in the standardized OrthoXML format, as e.g. generated by Inparanoid or OMA Standalone. PhyloProfile expects the NCBI taxonomy IDs to be present in the species tag in the XML as NCBITaxId like this:

<species name="Dipodomys ordii" NCBITaxId="10020">

OMA Standalone

By default, the output of OMA Standalone does not include the correct NCBITaxId (c.f. data/demo/oma_example.orthoxml) but rather gives these as <species name="Dipodomys_ordii" NCBITaxId="-1">. With scripts/ we provide a basic Python script to enable the use of OMA Standalone.

Besides the OrthoXML of OMA Standalone it only requires a simple, tab-separated mapping-file that maps the species names as generated by OMA Standalone to the NCBI Taxonomy ID. OMA Standalone uses the filenames of the protein sets you put into the DB folder as the species names, with Dipodomys_ordii.fa being transformed into the species name Dipodomys_ordii. An example mapping file should look like this:

$ head -n 3 data/demo/taxon_mapping_oma_orthoxml.csv
Dipodomys_ordii	10020
Mus_musculus	10090
Rattus_norvegicus	10116

To convert the OrthoXML of OMA Standalone to a PhyloProfile compatible OrthoXML you can simply run

./scripts/ -x data/demo/oma_example.orthoxml -m data/demo/taxon_mapping_oma_orthoxml.csv > data/demo/oma_example_phyloprofile_compatible.orthoxml

OMA Browser

If you prefer to use precalculated Hierarchical Orthologous Groups (HOGs) from the OMA Browser you can download your HOGs of interest right away from the commandline. To this end we provide scripts/ The input for this script is a list of OMA protein IDs (e.g. RATNO03710) or Uniprot IDs (e.g. P53_HUMAN). The types of IDs can be mixed.


./scripts/ -i RATNO03709 RATNO03710 RATNO03711 P53_HUMAN

will yield a single, merged OrthoXML files that contains the four HOGs for the corresponding proteins.

Your tool is missing?

Please get in touch! we are trying to support more orthology prediction tools right out of the box.

Demo data

In data/demo/ you can find some test data to see how the files should look like. In our Wiki you can find a walkthrough through the different files and how they relate to the different functions of PhyloProfile.


Any bug reports or comments, suggestions are highly appreciated. Please open an issue on GitHub or be in touch via email.


We would like to thank 1) Bastian for the great initial idea and his kind support, 2) Members of Ebersberger group for many valuable suggestions and …bug reports :)

Code of Conduct & Lisence

This tool is released with a Contributor Code of Conduct & under MIT lisence.


Vinh Tran