A look into reproducibility. Following a successful sensitization and introduction to concepts and tools used in Bioinformatics and Open science, we provided selected participants with projects to put into practice the skills they had learned. The Bioinformatics and Open Science Skills (BOSS) mini-project was a three-week event that gave the trainees a platform to collaborate by working and practising their skills. The main mini-projects were as follows:

  • Open science in Africa
  • Plant genome Analysis
  • Viral Analysis
  • Metagenomics Analysis
  • Open Science Research Data management handbook

Project selection

Reproducibility allows a continuation of work. Learning from what has already been done before presents a good opportunity to learn through practice. We sought to select papers whose data met FAIR principles and their methodologies could be reproduced to produce similar or close results. We came up with five projects for 20 participants to be divided into 5 groups. With help from our mentors and experts in various institutions, and following the interests expressed by our participants from previous surveys. We selected projects on plant and virus genomics, and metagenomics.  We also came up with two projects to focus on open science; the research data management handbook and the open science in Africa project. The open science in Africa project was an extension of the open science in Kenya project but focused on the data science aspect. With the projects at hand, we made the call for participants.

Git and GitHub recruitment and applications.

The best way to learn is to practice. We chose to use GitHub entirely for the mini-projects, starting with the applications. We created issues for each project and sent emails to our participants to comment on the projects they were most interested in. There were fewer applications a week after the call went out. We, therefore, created a google form for more applications as well. Week two saw a drastic increase in the number of applications mostly for the genomics projects, and more so for the plant genomics project which involved whole-genome assembly. The open science in Africa and research data management handbook projects had the least applications even toward the end of the application period. 

Selected participants were replied to via GitHub and via email as well. A few participants were moved from their preferred projects to the two projects with the least applicants to fill out the numbers to ensure each project had 5 participants. The core team members and experts were the mentors of the groups. 

Zoom and GitHub collaboration.

The selected participants came from countries including Kenya, Tanzania, Uganda and Ghana. The selection criteria were based on their experiences,  motivation, and availability. We held weekly meetings with all participants for updates, and more meetings for each project with the individual mentors. We had almost 100% consistent attendance during the weekly meetups. 

Mini project wins and fails.

This phase displayed the possibilities and challenges faced when working virtually and collaborating with different people from different regions with varying motivations. The participants arranged meetings outside the normal Friday weekly meetings we held with them during the projects. They had not taken entirely to GitHub for collaboration but managed to move together. Slack was a major part of the collaboration with each group having their own slack channel dedicated to project conversations. The three-week period was the biggest complaint from participants, reporting the limited time to figure out the project workings. This phase also showed the need for training on the use of high-performance clusters given that some participants had to play catch up with getting comfortable with using the platform. At the end of the phase, we had to strategize on how to conduct the remaining project and assess how best to ensure the completion of projects the next time we ran such an activity.  

Do you want to Hack??

Check out the BOSS mini-project GitHub repository and try out some of these projects. Reach out to us at for assistance or collaboration. These are amazing practice projects and we would be happy to assist anyone interested in skilling up in any way we can.