Project 2

Development of an online database and visualisation interface for livestock microbiome studies, with a focus on African datasets: LiMiDB (Livestock Microbiome DataBase)

Project objective

To implement and publish online a database for livestock microbiome studies (metagenomics), complemented with an interactive web interface to help with the visualisation and comparative analysis of such studies. We codename this project LiMiDB, for “Livestock Microbiome DataBase”.

Project leader

BecA-ILRI Hub

Dr Jean-Baka Domelevo Entfellner

BecA-ILRI Hub

Co-leader

John Innes Centre

Dr Oluwaseyi Shorinola

John Innes Centre

Researchers

Jomo Kenyatta University

Isaac Njaci

Jomo Kenyatta University
Jomo Kenyatta University of Agriculture and Technology

Mary Maranga

Jomo Kenyatta University of Agriculture and Technology
Egerton University

Edwin Murungi

Egerton University
Addis Ababa University

Helen Nigussie Aychegrew

Addis Ababa University
University of Nyala

Hassan Zackaria Ali Ishag

University of Nyala
Nelson Mandela African Institution of Science and Technology

Beatus Lyimo

Nelson Mandela African Institution of Science and Technology
Université Evangélique en Afrique

Bwihangane Birindwa Ahadi

Université Evangélique en Afrique
University of Ibadan

Osaiyuwu Osamede Henry

University of Ibadan

Project overview 

Metagenomics studies implement the sequencing and analysis of all fragments of DNA and/or RNA available in a given biological medium (e.g. skin tissue, gut or rumen sample, fecal material, leaf surface, soil sample, etc) and reveal genomic sequences belonging to a variety of micro-organisms. Recent studies have confirmed that these micro-organisms play an important role in facilitating biological processes in the host (e.g. digestion, nutrient absorption, immunology, etc) as well as the adaptation of the host to an adverse environment.

While metagenomics are a very promising and fast-developing field of biological sciences, they require complex workflows of downstream analysis and lead to many different types of quantitative results and graphical visualisations. This complicates the path from field sampling to proper analysis and interpretation of results.

Currently the vast majority of available metagenomics studies focus on human metagenomics, but the field also holds great promise for livestock, as better understanding of microbiomes could translate into innovations for better animal health, better breeding practices, reduced use of antibiotics, etc. Moreover, while metagenomics studies performed on material from Europe and America are more and more common, the adoption curve for this type of studies is slow among African scientists. We want to lift these barriers: we believe that the African scientific community (and, first and foremost, the alumni of the ABCF programme who were many to attempt such studies in the recent years) will benefit from using an online tool enabling them to upload their data and have a direct access to analysis results and visualisations, including tools for comparative analysis of different metagenomics studies. A comprehensive database of studies on livestock-related metagenomics studies will provide the data source for this online tool, with features enabling researchers to upload their dataset, either in unprocessed (raw Illumina reads) or semi-processed (BIOM files, abundance tables, etc) formats.

From the very beginning of the project implementation, care was taken to craft metadata categories and descriptions (in the design of the database schema) that allow for a seamless future extension of this tool to encompass metagenomics studies on domestic and wild animal, zoonotic diseases, crops, soil, water and environment. Ultimately, this database could form the primary support tool for all the metagenomics work carried out within the AfriqueOne-ASPIRE consortium (OneHealth paradigm).

Project outputs

  1. A comprehensive database of studies on livestock metagenomics studies.
  2. An online tool to query this database, visualise filtered results interactively and downloads various output tables and graphics.
  3. A module for that online tool to enable comparative analysis of metagenomics studies.
  4. A module for uploading new data, possibly in an unprocessed format (hence with a full analysis pipeline running in the background).

Project activities

  1. Communication with the ABCF alumni who have metagenomics sequence data, to explain the projects and get them to share their data
  2. Literature review to get a comprehensive list of non-ABCF African studies on livestock-related microbiomes and integration of the published datasets to the LiMiDB database
  3. Database schema design (MySQL) for LiMiDB
  4. Implementation of the data analysis and visualisation scripts (in R) using RMySQL to build queries to the database
  5. Development of the web interface to LiMiDB with Shiny and shinydashboard (two R packages)
  6. Implementation of the upstream data analysis workflow (shell scripting to wrap Qiime2 workflows) and integration thereof in the Shiny webapp