FOREWORD
The first third of 2021 is already over with the clock ticking towards the end of the H3ABioNet grant. In April H3ABioNet attended and participated in the 17th H3Africa consortium meeting which was again held virtually. Many H3ABioNet members hold key positions in the H3Africa working groups and several of the fellows gave presentations. The meeting provided an opportunity to get updates from the H3Africa projects, hear interesting talks from guest speakers, and to present our own updates. We helped to organize satellite courses and workshops before and after the meeting. Now we are working on the organization of the next H3ABioNet AGM and SAB meeting.
There has been a strong focus over the last few months on developing the timetable to fit all our planned online remote training courses into the coming year, the schedule is tight. The NGS course we are running with the Wellcome Connecting Science team is progressing with a few challenges associated with the large participant cohort and some technical limitations at sites, but these are being ironed out and we hope the participants will derive benefit from the course. This course is discussed further in the newsletter. The letter also features two articles that highlight women in science, including the article from our Training and Outreach Coordinator, Paballo Chauke, that was published after the Women in Data Science Africa event, and the acknowledgement of Tsaone Tamuhle’s L’Oréal-UNESCO Women in Science award. I hope you enjoy reading these and other stories.
Nicky Mulder
H3ABioNet PI
Reflections from Working on the H3ABioNet Website: Pros/Cons and Lessons Learnt
By: Wisdom A. Akurugu, PhD Student, University of Cape Town
have been involved with the activities of the H3ABioNet Consortium since my work as the Bioinformatician for the Noguchi Memorial Institute for Medical Research (NMIMR) Node in Ghana. When I look back from 2013 to date, there are many positives to reflect on while I’m yet to find the negatives. Having graduated with an MSc in Bioinformatics in 2013, joining the H3ABioNet consortium came in handy for a practical experience. It has always been exciting working for the consortium since then as I had the opportunities to attend working group meetings, workshops as well as the annual consortia meetings of H3ABioNet. I have positive recollection and great nostalgia when I reflect on my assistance in organising NMIMR’s node workshops and coordinating the IBT classroom activities back then. Another worthwhile reflection was my involvement in organising the monthly webinars that offered me a valuable platform to learn more about the discipline of bioinformatics and to build a social network of my future career. Working with the consortium has led to my current position as a PhD student at the central Node, University of Cape Town. Driven by my passion to pursue a career in Bioinformatics and Computational Biology, I approached Professor Nicola Mulder, PI of the consortium for any available opportunity to continue to assist with the activities of H3ABioNet. Thankfully, that moment came in 2018 and I have since been working for the consortium at the central node with an additional role as a node ambassador.
An important reminder is that H3ABioNet aims to develop bioinformatics capacity in Africa and specifically to enable genomics data analysis by H3Africa researchers. To achieve its purpose, H3ABioNet develops human capacity by conducting training and providing support for data analysis, and facilitating access to informatics infrastructure by developing or providing access to pipelines and tools for human, microbiome and pathogen genomic data analysis. To further establish its identity, disseminate these wonderful opportunities and communicate with the bioinformatics community, the consortium established and operates its website at H3ABioNet. I have been responsible for the management of some of the content of this site since joining the central node and which has been a great experience to share particularly with the use of the Joomla Content Management Tool.
Joomla is an open source content management system that helps you build powerful dynamic websites and applications. A web-based application that provides functionalities for multiple users with differing permission levels to manage all or a section of content, data or information of a website. It is flexible for creating, editing, archiving, publishing, collaborating on, reporting and distributing website content. Joomla can be used for corporate websites or portals, small business websites, magazines, newspapers, publications etc. With its intuitive interface, it helps you utilise its features and functionality to the best.
The core features of Joomla include user management which allows users to configure personal options while the media manager enables handling of media files or folders. Others include banner management, a searchable capability, content management, template management, integrated help system, system features, web services and a powerful extensibility.
Joomla is written in PHP and uses MySQL database to store the data while using object-oriented programming techniques. Basically, one does not need to learn any programming language to operate a Joomla website. It’s however, handy to have a working knowledge of HTML and CSS but most people only need to know drag-and-drop or how to fill in form settings in the backend. My experience indicates that to have great confidence and control of your website especially writing your own extensions, then you need a working knowledge in PHP, JavaScript, HTML, CSS, XML and SQL coding constructs.
Managing the content of the H3ABioNet website has been a great learning opportunity for me. My transition from a purely biological background (Biochemistry) to the application of computer science for the understanding of biology has really been a very slow process which is currently still ongoing. My basic programming lessons began with Python. Though not a wiz, I had some working knowledge to solve basic problems. I had no training in web development and its maintenance whatsoever. That posed some initial challenges for me working on the H3ABioNet website over a period of time. Perhaps, I have the fear of not being more aggressive and explorative enough, simply avoiding breaking down the site. Thankfully, one of our software developers, Mr. Mamana Mbiyavanga, who is one of the main architects of the site, was always available to assist. If there is anybody who dedicated time and space in getting me to appreciate the features and use of Joomla, mention must be made of Mamana.
Taking up a new role requires time and dedication to understand the concepts that define the area most especially developing the needed skills for the task. This requirement could be normal for everyone but I think I needed this badly. But this also came at a time I had just begun with my PhD studies which requires similar time and dedication in order to find my footing on the program. I have never combined working with studies. This I dare say was a test case for me. Those with this capability deserve commendation.
An important activity I undertook while managing the content of the website was the development of the new nodes’ pages. At the time, there was only one page that listed the various countries and under each a list of the PIs and their respective node’s names. There was also the consortium map which allowed a visitor to the page to find a node PI by clicking on the relevant node location. Quite substantial information on each node was missing. No node biography was available, individual members were not listed, no photos of members nor their skillsets. However, the REDCap survey had all these pieces of information sitting on the database thanks to Dr Sumir Panji and Mrs Katherine Johnston. It was therefore easy to pull these together for the new node pages. So in order to give the page a new look, the central node decided that we undertake that exercise and I had the responsibility to develop it.
One important reflection worthy of mention is the consensus building process on the framework to adopt. Again, with no web development experience nor artistic skills, I was confronted with producing a nice design of the pages outlook. I managed to put up a rough skeletal design and showed it to some members of the central node. I had very positive feedback especially from Mr. Mamana Mbiyavanga, Mrs Katherine Johnston and Dr. Sumir Panji, the Network manager, who all provided substantial input to the initial framework. The rest of the central node members also had a look and it finally got to the PI, Professor Nicola Mulder. A few additions were made and the final go ahead given. My experience back home was that I could most often walk into anyone’s office without a prior notice and discuss any issue. It turns out that that normally throws the person out of his or schedule causing an inconvenience and a stall of work. So building this consensus on the structure of the node pages had to be done via email and it apparently took longer than expected as most people had quite tight schedules. The questions I used to ask myself included, did these people get my email? Do they value what I did? Does it make sense at all? But with time and coupled with my own PhD studies experience, I got to really appreciate that one could be overwhelmed the whole day with loads of work and wouldn’t have enough time to respond to all emails. In the end we had an agreed and nice framework to build on.
One key feature of the pages is the dropdown list of the nodes. Though the Joomla tool has modules for most features, a dropdown module that suited us was not really available. The beauty of Joomla is a flexibility to build these modules and use. There again, my limitation on programming languages like CSS came up. I had to spend time to learn a few things about CSS and with the help of Mamana, we were able to get the desired dropdown list. Additionally, I needed to resize all the members’ photos to the required size while maintaining their quality. I managed to get a python code and with assistance was able to resize the photo including nodes logos and countries flags. I then got into putting everything together for the central node which became the template for all other nodes.
As I look back today and recall my experiences on managing the content of the website, I would say it has been great. It has introduced me to the concept of web development and the needed skillsets that one would have to acquire. It has given me a fair idea of the variety of programming languages and skillsets that I must develop in order to add value in the discipline of Bioinformatics. I’m most grateful to PI, Professor Nicola Mulder for this opportunity and the staff of H3BioNet at the central node for the support thus far.
I very much appreciate the opportunity to share this experience on content management of the H3ABioNet website and my general experience working with consortium members. Special thanks to you the reader.
L’Oréal-UNESCO Women in Science awards, sub-Saharan Africa young talents 2020
By: Tsaone Tamuhla, PhD student, University of Cape Town
I’m a third year PhD student (Bioinformatics) in the division of computational biology at the University of Cape Town (UCT), with a fellowship through H3ABioNet. I’m the node ambassador of the H3ABioNet Data Integration node and my node PI and PhD supervisor is Associate Professor Nicki Tiffin. It was A/Prof Nicki Tiffin, who first saw the award call and encouraged me to apply. Competition for these awards is very high and 2020 was no exception with over 300 applicants but only 20 awards (15 PhD and 5 Post-Docs), so I was not expecting to make it. Looking back at past recipients and at the 2020 cohort, it was truly a humbling experience to receive such recognition from L’Oréal-UNESCO. In research, we are often plagued by imposter syndrome, so being the recipient of an award that recognises academic and research excellence and exclusively seeks to elevate women in science has been a huge positive affirmation which has empowered me to truly appreciate the importance of my research in the grand scheme of things.
Women make up just 31% of researchers in Sub-Saharan Africa and only 3% of Nobel Prizes for Science have ever been awarded to women. I think the greatest accomplishments for women in science have been to keep going and striving to achieve and make a mark despite the many barriers. If you look at the achievements of women such as Wangari Muta Maathai, Marie Curie, Mae C. Jemison, Yvonne Clark and Rosalind Franklin to name a few, they are truly a testament to the brilliance, tenacity and resilience of women. Fortunately for us and future generations, we have organisations such as the L’Oréal Foundation, who in partnership with UNESCO are actively trying to remove some of those barriers and change the narrative of the involvement and achievements of women in science using platforms such as the women in science awards. Since winning the award I’ve been asked for advice and tips about the application process. Personally, I don’t believe there is a formula to a successful application but having undergone the process my biggest advice would be to be authentic. It’s ok to ask previous winners what they did, but your application is a story of your journey is science, and you should tell it in your own way. Lastly, to anyone looking to apply I say go for it, do not reject your own application by not submitting it.
Helpdesk Highlights – Get to Know the Experts
By: Zoe Gill, PhD student, University of Cape Town
Chip Imputation - Mamana Mbiyavanga
Tell us a bit about yourself:
My name is Mamana Mbiyavanga, a bioinformatics software developer in H3ABioNet central node at the University of Cape Town (UCT) in South Africa.
What does chip imputation entail?
Genotype imputation is a technique used in genome-wide association studies to compensate for the lack of comprehensive genomic coverage by genotyping arrays (i.e. H3Africa chip) and to deal with the differences in variants genotyped between different arrays when combining datasets. It is done by estimating missing genotypes from a denser pool of haplotypes, called a reference panel. It can effectively boost the power of signal detection in genome-wide association studies, integrate multi-studies for meta-analysis, and be applied in fine-mapping studies. The performance of genotype imputation is affected by many factors, including software, reference panel selection, sample size, and SNP density/sequencing coverage.
What type of questions can users ask in the data analysis - Chip Imputation category?
Some of the asked questions include:
- How do I get an account on the service?
- Is my data secure?
- What reference panel do I have access to?
- How do I download my results?
In your experience, what is most frequently asked with regards to this category?
What is the expected imputation performance or accuracy compared to other reference panels such as the 1000 Genomes project, Sanger imputation service African Genome Resources, Haplotype Reference Consortium, UK10K, TOPMED reference panels?
How to get started with Chip Imputation?
One gets started with Chip Imputation with the H3Africa Imputation Service (H3AIS) by simply requesting an account creation on https://impute.h3africa.org and then using the user-friendly web service to perform your imputation.
Are any online resources available where I can find more info on Chip Imputation?
The H3Africa Imputation Service (https://impute.h3africa.org) will provide all necessary information that would be needed regarding the Chip Imputation Helpdesk category.
Helpdesk Highlights – Get to know the experts
By: Zoe Gill, PhD student, University of Cape Town
Social Media Expert – Paballo Chauke
Tell us a bit about yourself:
I am a Training and Outreach Coordinator for H3ABIONET, based at the University of Cape of Town under the leadership of Professor Nicola Mulder. We are the central node meaning we get to liaise with all the other nodes in the continent. My job title is self-explanatory, I am tasked with working on H3ABIONET’s training such as workshops, hackathons and online/blended training such as Introduction to Bioinformatics as well as outreach. At the moment I am focusing on the outreach aspect of my job which is mainly about external work for promoting our project to our members, affiliates, funders, and general society. We attend science forums, create promotional materials such as banners and business cards, work on the website etc most importantly and easily accessible is the social media such as Facebook, Twitter, YouTube and others to promote H3ABIONET
What does being a Social Media expert entail?
It entails being in touch with people and also having professional and good soft communication skills, it is the awareness of what works for people to follow and making sure that our brand is visible, recognizable and influential. We use our social media to promote our publications, events, training, awards, opportunities, website, helpdesk. Social media is a powerful tool that can be strategically used for science communication and it entails a lot of thinking, strategizing and being on the ground.
What type of questions can users ask in the social media category?
Anything to do with our website and social media sites.
In your experience, what is most frequently asked with regards to this category?
People want to sign up for our website which is sweet but we do not allow for people to sign/register to use our website.
How to get started with social media?
I have given some training on this and I can share the slides with those that are interested, hit me up and I can assist you.
Data Science Needs Input from Women
By: Paballo Chauke, Training and Outreach Coordinator, University of Cape Town
Article originally published in the mail and guardian here: https://mg.co.za/opinion/2021-04-11-data-science-needs-input-from-women/
Being involved in the inaugural Women in Data Science (WiDS) Africa conference struck me as an opportunity to try to give credence to literary theorist and feminist critic Gayatri Chakravorty Spivak’s famous essay on colonialism and allow the “subaltern to speak”.
As a feminist ally, I passed the mic to some phenomenal women who are generally underpaid, underrepresented and unseen in the world of data science which, especially in Africa, is male-dominated.
Data has been referred to as the new oil because it is a valuable asset that has the potential to transform the way society operates. Data science is the interdisciplinary, complex employment of big data, artificial intelligence and advanced analytics and is a fast-growing field of science. It is increasingly being used to inform policy and decision-making in both government and the private sectors, but representation in the field remains unequal.
According to recent statistics, women make up a measly 15% of all data scientists globally and the World Economic Forum 2020 report found that only 26% of professionals employed in data and artificial intelligence in Africa are women.
Equal representation is crucial, however, because data is used to harness important information, often about people. Data has the potential to perpetuate harmful and oppressive biases; machines and robots are susceptible to learning and regurgitating human weaknesses and whims. As such, diversity, in terms of race, class, gender, sexuality, ability and other identity markers, is critical to the fair advancement of the field.
The tumultuous history that the world, and Africa in particular, has with science will only be changed from within. Inequality, expressed as racism, sexism and more has invariably been endorsed and promoted by scientists. For example, the myth that some racial groups are superior has propped up colonialism, apartheid, eugenics and genocide.
Science is often touted as an apolitical and ahistorical search for “objective knowledge” and understanding of our world. It follows a systematic methodology that is based on evidence. Even though organisations such as the Science Council provide quality assurance for scientists all over the world based on criteria such as objectivity and critical analysis, other scientists such as Keith Punch have argued that human beings are incapable of being objective — which means that science performed by any human is subjective.
And, regardless of the claims by many who want to continue to pretend otherwise, science is political, because it is in constant conversation with power in all spheres of society.
Furthermore, science is an organised knowledge community which has authority to affect lives whether positively or otherwise. Therefore this “neutral” science is, in Gramscian terms, propagated through hegemonic power and is normalised as common sense to uphold the status quo.
It is thus urgent that more women of colour cement themselves in this realm. The urgency with which data science needs transformation was echoed by women speakers and data scientists at the WiDS Africa conference, especially held on International Women’s Day, 8 March.
International Women’s Day commemorates the scientific, cultural, political, historical and socioeconomic achievements of women. It is also a focal point in the women’s rights movement, bringing crucial attention to issues such as gender inequality, encapsulated in the pay gap and the dearth of women in fields such as data science — issues that were highlighted at the much-referenced Beijing World Conference on Women in 1995.
But, 26 years later, not much has changed and researchers report that it could still take 163 years before gender equality is achieved.
WiDS Africa is part of the annual WiDS Worldwide conference organised by Stanford University, where about 150 events were hosted, featuring outstanding women doing outstanding work and which also aim to “encourage secondary school students to consider careers in data science, artificial intelligence, and related fields”.
I was tasked by Nicola Mulder, principal investigator at the Pan African Bioinformatics Network for H3Africa, to bring mainly women collaborators together for the local event.
According to the 2021 Women in Data Science report, women face myriad blockages in joining science, technology, engineering and mathematics fields, which include the absence of role models and mentors to encourage them to take these subjects and careers in these fields, as well as limited scholarships and finances to continue professional development.
In the WiDs Africa panel discussion on how the Covid-19 pandemic has affected their lives, Verena Ras, Chenai Chair, Rahab Wangari, Ruthbetha Kateule and Dominique Anderson echoed these views.
University of Cape Town vice-chancellor Mamokgethi Phakeng’s keynote speech addressed the relevance of women in data science.
I implore other men and feminist allies to give up their seat at the table, to stop being gatekeepers and actively create opportunities for women to join the science fraternity. We should all be feminists who dismantle patriarchal science through affirmative action, which, according to Ugandan author and lawyer Sylvia Tamale, “is a means of assuring substantive representation of a special group”, in this case women, and particularly black women.
We need to interrogate and highlight the limited representation of African women in data science (and other scientific fields) and resolve it soon before the rest of the world leaves us behind. Women’s views, nuanced experiences and knowledge are important additions to science to enrich and advance all humanity.
H3ABioNet and Wellcome Connecting Science team up to bring NGS to more Africans than ever before!
By: Verena Ras, Training and Outreach Coordinator, University of Cape Town
This year H3ABioNet teamed up with Wellcome Connecting Science to deliver their Next Generation sequencing Bioinformatics Training Course, to more African researchers and students than ever before. The goal – to get as many Africans trained in NGS data analysis using the H3ABioNet multiple-delivery-approach training model. An initial call saw around 40 classrooms apply (31 were eventually selected) to host the course locally, with over 2000 participants applying to take the course across these classrooms. We even saw many students from outside Africa apply! Since this was a pilot however, a smaller cohort needed to be selected resulting in roughly 400 participants being formally enrolled across the 31 classrooms. A whopping 100 staff members are also spread across the classrooms for local support.
The course kicked off on 13 April for all participants and we are swiftly moving through the 10 modules which cover topics ranging from quality control and sequencing technologies to variant calling and whole genome assembly.
The course marks yet another milestone for the multiple-delivery-approach model, drawing keen interest from international organisations as can be seen with this collaboration – hopefully the first of many. The pilot round of the course this year is a test… A test to see if we can collaborate with external organisations and re-package their expertly designed courses with a uniquely African twist. Should we be successful it paves the way for many other organisations to further develop this model and bring even more expertly designed courses to Africa.
What’s even better is that the model requires participation at all levels on a completely voluntary basis which means this incredible course (and all others run using this model) is 100% free for both local host classrooms and all their participants! Who said education does not come cheap!
We as H3ABioNet challenge all other providers to develop solutions similar to this and to bring quality education to those often forgotten.
Courses/Training
- Our flagship Introduction to Bioinformatics Training Course (IBT) has just closed this year’s round of participant applications. The course is set to start on 13 July for all participants! This year we have 54 classrooms spread across ~18 African countries.
- The first ever African, Next Generation Sequencing, Blended Learning Course run using H3ABioNet’s multiple-delivery-approach model started on 13 April and concluded on 15 June 2021. The course saw just over 400 participants enrol into this intensive 10 Module Course - the largest cohort ever trained within a single course on the continent.
- New FREE FutureLearn online course | Train the Trainer: Design Genomics and Bioinformatics Training | Learn how to train students, peers, professionals, and future experts in genomic techniques. Starts 21 June on Futurelearn: apply here.
- H3ABioNet will host a Research Data Management Plans Workshop from 22 - 25 June 2021. The course saw over 500 participants apply and hopes to deliver key introductory level data management training.
Grants and Awards
- The National Institute of Drug Abuse (NIDA) recently published the funding opportunity, “Leveraging Artificial Intelligence (AI) tools for Substance Use Disorders (SUD) drug discovery and development”. The non-dilutive funding is provided via the Small Business Innovation Research (SBIR) and Small Business Technology Transfer (STTR) funding programs. Funding ranges between $150K and $2.8M: 1) RFA-DA-22-019 Small Business Innovation Research (SBIR) 43/R44 | https://grants.nih.gov/grants/guide/rfa-files/RFA-DA-22-019.html AND 2) RFA-DA-22-021 Small Business Technology Transfer (STTR) R41/R42 | https://grants.nih.gov/grants/guide/rfa-files/RFA-DA-22-021.html. Application Due Date: August 10, 2021
- Launching Future Leaders in Global Health (LAUNCH) Research Training Program (D43 Clinical Trial Optional) funding opportunity. The purpose of this program is to provide opportunities for up to six consortia to develop or expand global health research training programs More information available here: https://grants.nih.gov/grants/guide/rfa-files/RFA-TW-21-004.html
- NIMH Funding Opportunity Announcement. Mental Health Research Awards for Investigators Early in their Career in Low and Middle-Income Countries. Find out more here https://grants.nih.gov/grants/guide/rfa-files/RFA-MH-21-120.html. Deadline 15 November 2021
Other Announcement/Events
- Wanting to run your own Bioinformatics workshops but don’t know where to start? H3ABioNet has developed an easy to use (step-by-step) support guide just for this reason. Find more information on our Training Guide here: https://h3abionet.org/categories/training/h3abionet-training-guide
- The International Society for Computational Biology is pleased to announce the HPC-AI Advisory Council (HPCAIAC) and National Supercomputing Centre (NSCC) Singapore 2021 APAC HPC-AI Competition. To become part of a team – register here: http://www.hpcadvisorycouncil.com/events/2021/APAC-AI-HPC/register.php
Credits:
creative commons