Bioinformatics in human health

Zimbabwe, like most African countries, battles with serious health challenges, particularly in infectious diseases such as the human immunodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS), malaria, and tuberculosis (TB) [17–20]. In addition, noncommunicable diseases are on the rise and contribute significantly to comorbidities. Researchers in Zimbabwe have started employing bioinformatics techniques to help unravel the genetic and environmental determinants for these diseases.

To better understand the HIV dynamics in vertical transmission, some research groups explored the diversity of HIV sequences from infected infants and various compartments from their mothers. In one study, genotypic analysis of HIV-1 envelope third variable loop sequences of infected but drug-naive women during pregnancy and their infected infants was carried out to predict virus coreceptor utilization in vertical transmission [21]. Using similar methods, other groups investigated the viral characteristics associated with high-risk genotypes of human papillomavirus that are associated with malignancy [22]. These studies represent an effort to use bioinformatics to produce data of clinical relevance for Zimbabwe and beyond.

Zimbabwean researchers have been at the forefront of developing the field of pharmacogenomics in Africa [23–26]. Spearheaded by the Departments of Biochemistry and Clinical Pharmacology at the UZ and AiBST, pharmacogenomics research has mostly focused on understanding the role of human and pathogen genetic variation in drug response, side effects, and therapeutic outcomes in the management of HIV, TB, and malaria. One of the most important outputs from these efforts has been the establishment of a biobank and a pharmacogenetic database for African populations, a resource developed to provide baseline understanding of pharmacogenetic variation for potential application toward tailoring drug treatment [24]. Further work has enabled elucidation of key markers, which are prevalent in African populations and are associated with side effects of the anti-retroviral drug efavirenz [25,26].

Zimbabwean scientists have started to embrace computational methods for the discovery of drug targets and for the identification of new disease markers to improve early diagnosis or develop new therapeutic strategies. For example, a study at AiBST used in vitro and in silico approaches to show that thiabendazole is a potent mechanism-based inhibitor of cytochrome P450 1A2 [27]. In the study, potential drug–drug interactions were also simulated. These and other simulation efforts are likely to increase given the workshops on computer aided drug discovery (CADD) conducted at the Zimbabwe Centre for High-Performance Computing (ZCHPC) in 2016.

Local scientists have not actively pursued high-throughput functional genomics approaches such as proteomic profiling for mechanism-based drug discovery and drug repurposing. This is despite the fact that gas chromatography–mass spectrometers (GC-MS) and liquid chromatography–mass spectrometers (LC-MS) can be found at research institutions such as AiBST, National Microbiology Reference Laboratory, and Standards Association of Zimbabwe Laboratories. This is mainly because there are few researchers with sufficient knowledge of proteomics (or other functional genomics approaches such as transcriptomics and metabolomics) to carry out research encompassing “omics”-driven objectives. Also, the organizations with the GC-MS and LC-MS generally do not have technicians with expertise in proteomics, and thus there are no optimized protocols and data analysis pipelines.

Bioinformatics in food security

Zimbabwe produces a large variety of agricultural products meant for food, such as maize, sorghum, groundnuts, fruit, and vegetables, in addition to livestock and poultry. Generally, bioinformatics can be used to efficiently gain access to “omics” data available in the publicly available repositories and to make available the data to research scientists involved in animal or crop breeding. For example, local scientists can employ genomics analysis to aid in the assessment of genetic diversity and identification of important genetic traits for the production of crops and livestock with desirable traits including high yield and disease resistance [7,8]. However, there is little bioinformatics activity in that direction in Zimbabwe. Research institutions such as universities, the Scientific and Industrial Research and Development Centre, the International Maize and Wheat Improvement Centre (CIMMYT), and the International Crops Research Institute for the Semi-Arid Tropics, are, however, well positioned to adopt bioinformatics techniques owing to the regional and international collaborative nature of their research projects. The following examples illustrate how such collaborations are important.

1. Researchers at CIMMYT, working with their international collaborators, conducted a metabolomic profiling study of maize leaves to investigate the effects of drought, heat, and combined stress on grain yield in field crops [28]. This kind of work is important in the development of abiotic stress-resistant and/or tolerant cultivars, thereby improving food security.
2. In the livestock sector, a genome-wide association study of the Zimbabwean goat, pigs, and cattle has been done by the South African Biotechnology Research Institute under the Agricultural Research Council in collaboration with the UZ [29].

Metagenomics studies

Some metagenomic efforts are being made at some universities such as CUT and National University of Science and Technology (NUST), with the aim of improving environmental sustainability and the bioprocessing industry. Among other works, the NUST group presented the whole-genome shotgun metagenome sequences of the greater kudu (Tragelaphus strepsiceros) rumen digesta revealing its diverse microbial community and some novel hydrolytic enzymes [30]. While all the preliminary wet laboratory procedures, including the metagenomic DNA extraction, were carried out at NUST, subsequent steps were carried out at international institutes. Whole-genome sequencing of the metagenomic DNA was performed at Inqaba Biotech Laboratories in Pretoria, South Africa, and the sequences were analyzed at the European Bioinformatics Institute of the European Molecular Biology Laboratory. In addition, a research group at CUT, working with South African collaborators, performed functional screening of the metagenome of the hindgut bacterial symbionts of a termite, Trinervitermes trinervoides, to discover open reading frames for 25 cellulases and hemicellulases [31]. These metagenomic efforts in the country have biotechnological significance because they may lead to, among other things, the isolation of novel enzymes such as cellulases, which can be used in the production of biofuels.

Bioinformatics education

Scientific networking and collaborations

Networking among local scientists presents potential opportunities for peer support, collaborative research, and effective postgraduate training of students among bioinformatics researchers. One of the reasons for poor networking among Zimbabwean scientists is that some institutions do not have comprehensive websites, and potential collaborators find it difficult to identify groups to work with. This puts many as yet unknown early career scientists at a disadvantage. We have summarized (Table 1) the institutions pursuing bioinformatics-related research. However, most of the research findings from early career scientists are yet to be published in peer-reviewed journals. We hope that publishing this list will encourage networking and collaborations among local scientists, as well as with the international community.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Research activities currently underway in Zimbabwean institutions.

https://doi.org/10.1371/journal.pcbi.1006480.t001

Following the first Zimbabwe Bioinformatics Symposium, the NBA made efforts to establish the Bioinformatics Consortium of Zimbabwe (BCZ), which aimed at (1) continually promoting networking among bioinformaticians, (2) promoting the exchange of ideas and resources, and (3) facilitating local and international collaborations. Through this consortium, an application was submitted and accepted for the Human Health Node through the H3ABioNet. With this encouraging development, there is a great opportunity for increased networking and membership of the BCZ.

We note that only a handful of researchers from Zimbabwe have joined the International Society for Computational Biology (ISCB) (https://www.iscb.org) or ASBCB. These societies provide an excellent forum for the interaction of researchers in the area of bioinformatics, which is highly beneficial to the members. Nonparticipation in the ISCB and ASBCB is probably due to the slow development of bioinformatics research in the country coupled with poor funding opportunities for local researchers.

Proposed strategy for implementing bioinformatics at national level

We are confident that a clear national road map aimed at developing independent and dedicated research and support structures as demonstrated by the South African experience [9] will accelerate implementation of bioinformatics research in Zimbabwe. Therefore, we propose that we devise strategies as summarized in Fig 2.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Proposed strategies for promoting bioinformatics in Zimbabwe.

In the short and medium terms, formation of research groups and clusters is essential while working toward the development of a green paper in the long run. In education and/or training, focus should be placed on workshops in the short term followed by introduction of MSc and PhD programs in the medium and long term. See main text for details.

https://doi.org/10.1371/journal.pcbi.1006480.g002

Short-term strategies

In the short term, the BCZ should call a meeting of all Zimbabwean bioinformatics researchers, educators, industry leaders, and government policy makers to discuss bioinformatics resources and opportunities. Such a situation would allow for the identification of national priorities for bioinformatics research. The specific research areas chosen should be in line with the developmental needs of the country and be feasible in terms of the country’s resources. Although there was an interaction of local researchers in the first Zimbabwe Bioinformatics Symposium held in Harare in 2016, the major focus was to present past and present research results by local and international scientists. The BCZ, which should adopt a more transparent membership recruitment policy, should drive the development of a roadmap and would be expected to review progress at least annually through research symposia. In that regard, the BCZ needs to develop a clear administration structure that should be tasked with coming up with a constitution and a website to facilitate the creation and/or editing of profiles of bioinformatics research groups, thus helping to publicize bioinformatics initiatives in Zimbabwe. In addition, the BCZ would be expected to provide ongoing support to members through web-based portals.

During the development of the national roadmap, an International Advisory Committee (IAC) needs to be formed, which would be responsible for providing guidance on the planning and implementation of the project as a whole. The IAC should include members from the region and the international community—thus combining the knowledge of the peculiarities in the region and the more objective points of view from the international community. For the IAC to be effective, it should be composed of members with no conflicts of interest regarding the development of bioinformatics in Zimbabwe.

To address the critical skills shortage, the short-term strategy would be giving focus to seminars, conferences, and workshops that are coordinated at a national level and tailored to meet the training requirements of postgraduate students, academics, and relevant industry players. Given that bioinformatics is an interdisciplinary field, postgraduate students and researchers can be drawn from a broad range of disciplines including biology, chemistry, statistics, and computer science. For beginners, the courses may be designed to teach participants important skills such as coding for biology, biological databases, data mining, etc. More advanced topics—including systems biology, proteomics, metabolomics, and structural bioinformatics—should be considered for those who are already familiar with the basic introductory topics. Prioritizing short courses at this stage will be important to ensure that research students become acquainted with the latest advancements in bioinformatics and thus design their research projects accordingly. Involvement of the government and industry is crucial because these can help provide postgraduate students funding for attending international conferences and training events that could be a good means of quickly getting the bioinformatics community active. Mentoring and training of trainers will be crucial for building capacity and developing critical mass of individuals with bioinformatics skills. In addition, the BCZ needs to develop an online platform for sharing educational and/or training material, code, and bioinformatics protocols. These measures will ensure easy access to information and speed up developments in the field.

Medium-term strategies

Once the stage has been set for the national coordination of bioinformatics activities, these need to be strengthened by the formation of research clusters. Because bioinformatics is inherently interdisciplinary in nature, involvement of computer scientists, statisticians, etc., is crucial. Formation of multidisciplinary research clusters should make it easier to procure research funding from local sources and competitive international sources. Research clusters should be supported at a national level by coordinating and facilitating access to infrastructure facilities. At present, many local research groups work in isolation, and some are not aware of where to access resources locally, thus contributing to a poor publication record. Ongoing research programs such as those highlighted in Table 1 could be used as the basis for development of specific bioinformatics skills.

It can be argued that the establishment of an MSc (or equivalent postgraduate qualification) in bioinformatics is urgent in Zimbabwe. One way to do this, as a starting point, would be to identify one university and support it to set up a national bioinformatics training and resource institute. The institute could house the majority of bioinformaticians available in the country and be key in the training of both students and staff and help in local bioinformatics research activities. The institute will also work closely with the BCZ to make sure that no researcher and/or research is excluded and unrecognized. It is refreshing to note that CUT has started efforts to draft regulations for MSc Bioinformatics degree programs by coursework in collaboration with H3ABioNet. Associating with H3ABioNet is ideal because it helps one benefit from the experience of setting up an MSc Bioinformatics program in an African setting to address African needs. Also, it helps to address the skills shortage for lecturers because the H3ABioNet has coordination strategies for sharing human resources from distant universities, including via online education. CUT has also signed a memorandum of understanding with AiBST in an arrangement that would see CUT offering an MSc in Genomics and Precision Medicine starting in 2019.

Long-term strategies

In the long term, it should be possible that a national green paper on bioinformatics be produced and agreed policies implemented. This can be achieved by involving the government from an early stage in defining the priorities for support. For example, given the burden of both communicable and noncommunicable disease in Africa, all stakeholders could agree that health is a national priority for bioinformatics support. Tackling emerging risks of viral infections and drug resistance can be addressed by applying bioinformatics to understand pathogen genomes, to follow resistance patterns, and for identifying drug targets.

With the background work done, the culmination of education and training efforts should see significant registrations for PhD studies and postdoctoral training. The importance of doctoral-level training at a local level for any economy need not be over-emphasized. Locally trained graduates would help tackle local problems more effectively, either during their training or after graduation. Indeed, the publication record for the country would significantly improve. The government can play an important role in supporting the PhD studies financially.

Conclusion

Despite limited funding sources, we find that some bioinformatics research efforts are being made by several groups involved in biomedical research, CADD, agriculture, and biotechnology. However, the most active areas are those involved in human health. Most of the biomedical research in Zimbabwe is being done by a few well-established institutions with links and partnerships to established institutions in the region or overseas. The research focuses mainly on the primary diseases of poverty (TB, malaria, HIV/AIDS, and other opportunistic pathogens). However, the national publication record is still low when compared to some African countries that have embraced bioinformatics, mainly due to poor funding, poor networking, and inadequate training opportunities. Solving these problems would require coordination at a national level.

We have proposed strategies for stimulating increased activities in bioinformatics in Zimbabwe. These strategies include the recognition of the BCZ that was established in 2016 to lead the formulation and implementation of the roadmap to bioinformatics growth and coordinate national activities, among other goals. In order for the BCZ to be more viable internationally, it should be affiliated with international platforms such as the ISCB and the ASBCB. Clearly, the envisaged responsibilities of the BCZ are serious and require an effective administrative structure. An effective BCZ is important because it can aid government policy makers in mapping the direction of science and technology research in the country.

For a sustainable implementation of bioinformatics in the country, we suggest that the country aim to become a data producer and bioinformatics service provider on a specific and firm set of limited domains rather than being an unfocused bioinformatics user in a broad range of disparate fields. Therefore, at this stage, the country should aim to operate within a targeted node in a pan-African network. In our view, although international linkages are important, the commitment to bioinformatics capacity development by the government and industry players would be key to the success of the bioinformatics endeavors in the country.