Using a multiple-delivery-mode training approach to develop local capacity and infrastructure for advanced bioinformatics in Africa

Abstract

With more microbiome studies being conducted by African-based research groups, there is an increasing demand for knowledge and skills in the design and analysis of microbiome studies and data. However, high-quality bioinformatics courses are often impeded by differences in computational environments, complicated software stacks, numerous dependencies, and versions of bioinformatics tools along with a lack of local computational infrastructure and expertise. To address this, H3ABioNet developed a 16S rRNA Microbiome Intermediate Bioinformatics Training course, extending its remote classroom model. The course was developed alongside experienced microbiome researchers, bioinformaticians, and systems administrators, who identified key topics to address. Development of containerised workflows has previously been undertaken by H3ABioNet, and Singularity containers were used here to enable the deployment of a standard replicable software stack across different hosting sites. The pilot ran successfully in 2019 across 23 sites registered in 11 African countries, with more than 200 participants formally enrolled and 106 volunteer staff for onsite support. The pulling, running, and testing of the containers, software, and analyses on various clusters were performed prior to the start of the course by hosting classrooms. The containers allowed the replication of analyses and results across all participating classrooms running a cluster and remained available posttraining ensuring analyses could be repeated on real data. Participants thus received the opportunity to analyse their own data, while local staff were trained and supported by experienced experts, increasing local capacity for ongoing research support. This provides a model for delivering topic-specific bioinformatics courses across Africa and other remote/low-resourced regions which overcomes barriers such as inadequate infrastructures, geographical distance, and access to expertise and educational materials.

Fig 1. Growth in participation within the H3ABioNet Introduction to Bioinformatics Training Course (IBT) over the period of 2016–2020.

Scales have been coloured to match the colours of their associated graphs.

Fig 2. Diagrammatic depiction of the multiple-delivery-mode training model as it was applied within the H3ABioNet 16S rRNA Intermediate Bioinformatics Training course during 2019.

The figure illustrates the 3 major modalities employed within the course, which are OER (Open Educational Resources), face-to-face learning, and online/distance-based learning. We also depict the staffing structure.

Fig 3. Simplified schematic representation of the 16S rRNA analysis workflow.

The complete TADA workflow is accessible on GitHub: https://github.com/h3abionet/TADA.

Fig 4. Responses from feedback surveys for randomly selected questions to illustrate participant satisfaction with the H3ABioNet 16S rRNA Microbiome Intermediate Bioinformatics Training course (https://www.h3abionet.org/training/int-bt).

Surveys were all administered within the learning management system Vula (http://vula.uct.ac.za). These summaries were compiled from a total of 93 responses across all questions.