Machine learning approaches classify clinical malaria outcomes based on haematological parameters.

BACKGROUND: Malaria is still a major global health burden, with more than 3.2 billion people in 91 countries remaining at risk of the disease. Accurately distinguishing malaria from other diseases, especially uncomplicated malaria (UM) from non-malarial infections (nMI), remains a challenge. Furthermore, the success of rapid diagnostic tests (RDTs) is threatened by Pfhrp2/3 deletions and decreased sensitivity at low parasitaemia. Analysis of haematological indices can be used to support the identification of possible malaria cases for further diagnosis, especially in travellers returning from endemic areas. As a new application for precision medicine, we aimed to evaluate machine learning (ML) approaches that can accurately classify nMI, UM, and severe malaria (SM) using haematological parameters. METHODS: We obtained haematological data from 2,207 participants collected in Ghana: nMI (n = 978), SM (n = 526), and UM (n = 703). Six different ML approaches were tested, to select the best approach. An artificial neural network (ANN) with three hidden layers was used for multi-classification of UM, SM, and uMI. Binary classifiers were developed to further identify the parameters that can distinguish UM or SM from nMI. Local interpretable model-agnostic explanations (LIME) were used to explain the binary classifiers. RESULTS: The multi-classification model had greater than 85% training and testing accuracy to distinguish clinical malaria from nMI. To distinguish UM from nMI, our approach identified platelet counts, red blood cell (RBC) counts, lymphocyte counts, and percentages as the top classifiers of UM with 0.801 test accuracy (AUC = 0.866 and F1 score = 0.747). To distinguish SM from nMI, the classifier had a test accuracy of 0.96 (AUC = 0.983 and F1 score = 0.944) with mean platelet volume and mean cell volume being the unique classifiers of SM. Random forest was used to confirm the classifications, and it showed that platelet and RBC counts were the major classifiers of UM, regardless of possible confounders such as patient age and sampling location. CONCLUSION: The study provides proof of concept methods that classify UM and SM from nMI, showing that the ML approach is a feasible tool for clinical decision support. In the future, ML approaches could be incorporated into clinical decision-support algorithms for the diagnosis of acute febrile illness and monitoring response to acute SM treatment particularly in endemic settings.

Stabilization of RDT target antigens present in dried Plasmodium falciparum-infected samples for validating malaria rapid diagnostic tests at the point of care.

BACKGROUND: Malaria rapid diagnostic tests (RDTs) are a great achievement in implementation of parasite based diagnosis as recommended by World Health Organization. A major drawback of RDTs is lack of positive controls to validate different batches/lots at the point of care. Dried Plasmodium falciparum-infected samples with the RDT target antigens have been suggested as possible positive control but their utility in resource limited settings is hampered by rapid loss of activity over time. METHODS: This study evaluated the effectiveness of chemical additives to improve long term storage stability of RDT target antigens (HRP2, pLDH and aldolase) in dried P. falciparum-infected samples using parasitized whole blood and culture samples. Samples were treated with ten selected chemical additives mainly sucrose, trehalose, LDH stabilizer and their combinations. After baseline activity was established, the samples were air dried in bio-safety cabinet and stored at room temperatures (~ 25 °C). Testing of the stabilized samples using SD Bioline, BinaxNOW, CareStart, and First Response was done at intervals for 53 weeks. RESULTS: Stability of HRP2 at ambient temperature was reported at 21-24 weeks while that of PAN antigens (pLDH and aldolase) was 2-18 weeks of storage at all parasite densities. The ten chemical additives increased the percentage stability of HRP2 and PAN antigens. Sucrose alone and its combinations with Alsever's solution or biostab significantly increased stability of HRP2 by 56% at 2000 p/µL (p < 0.001). Trehalose and its combinations with biostab, sucrose or glycerol significantly increased stability of HRP2 by 57% (p < 0.001). Unlike sucrose, the stability of the HRP2 was significantly retained by trehalose at lower concentrations (500, and 200 p/µL). Trehalose in combination biostab stabilizer increased the percentage stability of PAN antigens by 42, and 32% at 2000 and 500 p/µL respectively (p < 0.01). This was also the chemical combination with the shortest reconstitution time (~ < 20 min). CONCLUSIONS: These findings confirm that stabilizing RDT target antigens in dried P. falciparum-infected samples using chemical additives provides field-stable positive controls for malaria RDTs.

Comparative evaluation of fluorescent in situ hybridization and Giemsa microscopy with quantitative real-time PCR technique in detecting malaria parasites in a holoendemic region of Kenya.

BACKGROUND: Early and accurate diagnosis of malaria is important in treatment as well as in the clinical evaluation of drugs and vaccines. Evaluation of Giemsa-stained smears remains the gold standard for malaria diagnosis, although diagnostic errors and potential bias estimates of protective efficacy have been reported in practice. Plasmodium genus fluorescent in situ hybridization (P-Genus FISH) is a microscopy-based method that uses fluorescent labelled oligonucleotide probes targeted to pathogen specific ribosomal RNA fragments to detect malaria parasites in whole blood. This study sought to evaluate the diagnostic performance of P-Genus FISH alongside Giemsa microscopy compared to quantitative reverse transcription polymerase chain reaction (qRT-PCR) in a clinical setting. METHOD: Five hundred study participants were recruited prospectively and screened for Plasmodium parasites by P-Genus FISH assay, and Giemsa microscopy. The microscopic methods were performed by two trained personnel and were blinded, and if the results were discordant a third reading was performed as a tie breaker. The diagnostic performance of both methods was evaluated against qRT-PCR as a more sensitive method. RESULTS: The number of Plasmodium positive cases was 26.8% by P-Genus FISH, 33.2% by Giemsa microscopy, and 51.2% by qRT-PCR. The three methods had 46.8% concordant results with 61 positive cases and 173 negative cases. Compared to qRT-PCR the sensitivity and specificity of P-Genus FISH assay was 29.3 and 75.8%, respectively, while microscopy had 58.2 and 93.0% respectively. Microscopy had a higher positive and negative predictive values (89.8 and 68.0% respectively) compared to P-Genus FISH (56.0 and 50.5%). In overall, microscopy had a good measure of agreement (76%, k = 0.51) compared to P-Genus FISH (52%, k = 0.05). CONCLUSION: The diagnostic performance of P-Genus FISH was shown to be inferior to Giemsa microscopy in the clinical samples. This hinders the possible application of the method in the field despite the many advantages of the method especially diagnosis of low parasite density infections. The P-Genus assay has great potential but application of the method in clinical setting would rely on extensive training of microscopist and continuous proficiency testing.

Evaluation of a laboratory quality assurance pilot programme for malaria diagnostics in low-transmission areas of Kenya, 2013.

BACKGROUND: One objective of the Kenya National Malaria Strategy 2009-2017 is scaling access to prompt diagnosis and effective treatment. In 2013, a quality assurance (QA) pilot was implemented to improve accuracy of malaria diagnostics at selected health facilities in low-transmission counties of Kenya. Trends in malaria diagnostic and QA indicator performance during the pilot are described. METHODS: From June to December 2013, 28 QA officers provided on-the-job training and mentoring for malaria microscopy, malaria rapid diagnostic tests and laboratory QA/quality control (QC) practices over four 1-day visits at 83 health facilities. QA officers observed and recorded laboratory conditions and practices and cross-checked blood slides for malaria parasite presence, and a portion of cross-checked slides were confirmed by reference laboratories. RESULTS: Eighty (96%) facilities completed the pilot. Among 315 personnel at pilot initiation, 13% (n = 40) reported malaria diagnostics training within the previous 12 months. Slide positivity ranged from 3 to 7%. Compared to the reference laboratory, microscopy sensitivity ranged from 53 to 96% and positive predictive value from 39 to 53% for facility staff and from 60 to 96% and 52 to 80%, respectively, for QA officers. Compared to reference, specificity ranged from 88 to 98% and negative predictive value from 98 to 99% for health-facility personnel and from 93 to 99% and 99%, respectively, for QA officers. The kappa value ranged from 0.48-0.66 for facility staff and 0.57-0.84 for QA officers compared to reference. The only significant test performance improvement observed for facility staff was for specificity from 88% (95% CI 85-90%) to 98% (95% CI 97-99%). QA/QC practices, including use of positive-control slides, internal and external slide cross-checking and recording of QA/QC activities, all increased significantly across the pilot (p < 0.001). Reference material availability also increased significantly; availability of six microscopy job aids and seven microscopy standard operating procedures increased by a mean of 32 percentage points (p < 0.001) and 38 percentage points (p < 0.001), respectively. CONCLUSIONS: Significant gains were observed in malaria QA/QC practices over the pilot. However, these advances did not translate into improved accuracy of malaria diagnostic performance perhaps because of the limited duration of the QA pilot implementation.

Leveraging wastewater-based epidemiology to monitor the spread of neglected tropical diseases in African communities.

Neglected tropical diseases continue to cause a significant burden worldwide, with Africa accounting for more than one-third of the global burden. Over the past decade, progress has been made in eliminating, controlling, and eradicating these diseases in Africa. By December 2022, 47 out of 54 African countries had eliminated at least one neglected tropical disease, and more countries were close to achieving this milestone. Between 2020 and 2021, there was an 80 million reduction in people requiring intervention. However, continued efforts are needed to manage neglected tropical diseases and address their social and economic burden, as they deepen marginalisation and stigmatisation. Wastewater-based epidemiology involves analyzing wastewater to detect and quantify biomarkers of disease-causing pathogens. This approach can complement current disease surveillance systems in Africa and provide an additional layer of information for monitoring disease spread and detecting outbreaks. This is particularly important in Africa due to limited traditional surveillance methods. Wastewater-based epidemiology also provides a tsunami-like warning system for neglected tropical disease outbreaks and can facilitate timely intervention and optimised resource allocation, providing an unbiased reflection of the community's health compared to traditional surveillance systems. In this review, we highlight the potential of wastewater-based epidemiology as an innovative approach for monitoring neglected tropical disease transmission within African communities and improving existing surveillance systems. Our analysis shows that wastewater-based epidemiology can enhance surveillance of neglected tropical diseases in Africa, improving early detection and management of Buruli ulcers, hookworm infections, ascariasis, schistosomiasis, dengue, chikungunya, echinococcosis, rabies, and cysticercosis for better disease control.

Non-Coding RNAs Potentially Involved in Pyrethroid Resistance of Anopheles funestus Population in Western Kenya.

Backgrounds: The resurgence of Anopheles funestus, a dominant vector of human malaria in western Kenya was partly attributed to insecticide resistance. However, evidence on the molecular basis of pyrethroid resistance in western Kenya is limited. Noncoding RNAs (ncRNAs) form a vast class of RNAs that do not code for proteins and are ubiquitous in the insect genome. Here, we demonstrated that multiple ncRNAs could play a potential role in An. funestusresistance to pyrethroid in western Kenya. Materials and Methods: Anopheles funestus mosquitoes were sampled by aspiration methods in Bungoma, Teso, Siaya, Port Victoria and Kombewa in western Kenya. The F1 progenies were exposed to deltamethrin (0.05%), permethrin (0.75%), DDT (4%) and pirimiphos-methyl (0.25%) following WHO test guidelines. A synergist assay using piperonyl butoxide (PBO) (4%) was conducted to determine cytochrome P450s' role in pyrethroid resistance. RNA-seq was conducted on a combined pool of specimens that were resistant and unexposed, and the results were compared with those of the FANG susceptible strain. This approach aimed to uncover the molecular mechanisms underlying pyrethroid resistance. Results: Pyrethroid resistance was observed in all the sites with an average mortality rate of 57.6%. Port Victoria had the highest level of resistance to permethrin (MR=53%) and deltamethrin (MR=11%) pyrethroids. Teso had the lowest level of resistance to permethrin (MR=70%) and deltamethrin (MR=87%). Resistance to DDT was observed only in Kombewa (MR=89%) and Port Victoria (MR=85%). A full susceptibility to P-methyl (0.25%) was observed in all the sites. PBO synergist assay revealed high susceptibility (>98%) to the pyrethroids in all the sites except for Port Victoria (MR=96%, n=100). Whole transcriptomic analysis showed that most of the gene families associated with pyrethroid resistance comprised non-coding RNAs (67%), followed by imipenemase (10%), cytochrome P450s (6%), cuticular proteins (5%), olfactory proteins (4%), glutathione S-transferases (3%), UDP-glycosyltransferases (2%), ATP-binding cassettes (2%) and carboxylesterases(1%). Conclusions: This study unveils the molecular basis of insecticide resistance in An. funestus in western Kenya, highlighting for the first time the potential role of non-coding RNAs in pyrethroid resistance. Targeting non-coding RNAs for intervention development could help in insecticide resistance management.

Drug resistance and vaccine target surveillance of Plasmodium falciparum using nanopore sequencing in Ghana.

Malaria results in over 600,000 deaths annually, with the highest burden of deaths in young children living in sub-Saharan Africa. Molecular surveillance can provide important information for malaria control policies, including detection of antimalarial drug resistance. However, genome sequencing capacity in malaria-endemic countries is limited. We designed and implemented an end-to-end workflow to detect Plasmodium falciparum antimalarial resistance markers and diversity in the vaccine target circumsporozoite protein (csp) using nanopore sequencing in Ghana. We analysed 196 clinical samples and showed that our method is rapid, robust, accurate and straightforward to implement. Importantly, our method could be applied to dried blood spot samples, which are readily collected in endemic settings. We report that P. falciparum parasites in Ghana are mostly susceptible to chloroquine, with persistent sulfadoxine-pyrimethamine resistance and no evidence of artemisinin resistance. Multiple single nucleotide polymorphisms were identified in csp, but their significance is uncertain. Our study demonstrates the feasibility of nanopore sequencing for malaria genomic surveillance in endemic countries.

Malaria artemisinin resistance: an extracellular vesicles export hypothesis.

Plasmodium falciparum causes malaria, and its resistance to artemisinin (ART) - a drug used for managing malaria - threatens to interfere with the effective control of malaria. ART resistance (ARTr) is driven by increased tolerance to oxidative stress and reduced haemoglobin trafficking to the food vacuole. We discuss how extracellular vesicles (EVs) may play a role in developing ARTr.

Detection of SARS-CoV-2 intra-host recombination during superinfection with Alpha and Epsilon variants in New York City.

Recombination is an evolutionary process by which many pathogens generate diversity and acquire novel functions. Although a common occurrence during coronavirus replication, detection of recombination is only feasible when genetically distinct viruses contemporaneously infect the same host. Here, we identify an instance of SARS-CoV-2 superinfection, whereby an individual was infected with two distinct viral variants: Alpha (B.1.1.7) and Epsilon (B.1.429). This superinfection was first noted when an Alpha genome sequence failed to exhibit the classic S gene target failure behavior used to track this variant. Full genome sequencing from four independent extracts reveals that Alpha variant alleles comprise around 75% of the genomes, whereas the Epsilon variant alleles comprise around 20% of the sample. Further investigation reveals the presence of numerous recombinant haplotypes spanning the genome, specifically in the spike, nucleocapsid, and ORF 8 coding regions. These findings support the potential for recombination to reshape SARS-CoV-2 genetic diversity.

Genetic diversity of SARS-CoV-2 infections in Ghana from 2020-2021.

The COVID-19 pandemic is one of the fastest evolving pandemics in recent history. As such, the SARS-CoV-2 viral evolution needs to be continuously tracked. This study sequenced 1123 SARS-CoV-2 genomes from patient isolates (121 from arriving travellers and 1002 from communities) to track the molecular evolution and spatio-temporal dynamics of the SARS-CoV-2 variants in Ghana. The data show that initial local transmission was dominated by B.1.1 lineage, but the second wave was overwhelmingly driven by the Alpha variant. Subsequently, an unheralded variant under monitoring, B.1.1.318, dominated transmission from April to June 2021 before being displaced by Delta variants, which were introduced into community transmission in May 2021. Mutational analysis indicated that variants that took hold in Ghana harboured transmission enhancing and immune escape spike substitutions. The observed rapid viral evolution demonstrates the potential for emergence of novel variants with greater mutational fitness as observed in other parts of the world.

High-throughput genotyping assays for identification of glycophorin B deletion variants in population studies.

Glycophorins are the most abundant sialoglycoproteins on the surface of human erythrocyte membranes. Genetic variation in glycophorin region of human chromosome 4 (containing GYPA, GYPB, and GYPE genes) is of interest because the gene products serve as receptors for pathogens of major public health interest, including Plasmodiumsp., Babesiasp., Influenza virus, Vibrio cholerae El Tor Hemolysin, and Escherichia coli. A large structural rearrangement and hybrid glycophorin variant, known as Dantu, which was identified in East African populations, has been linked with a 40% reduction in risk for severe malaria. Apart from Dantu, other large structural variants exist, with the most common being deletion of the whole GYPB gene and its surrounding region, resulting in multiple different deletion forms. In West Africa particularly, these deletions are estimated to account for between 5 and 15% of the variation in different populations, mostly attributed to the forms known as DEL1 and DEL2. Due to the lack of specific variant assays, little is known of the distribution of these variants. Here, we report a modification of a previous GYPB DEL1 assay and the development of a novel GYPB DEL2 assay as high-throughput PCR-RFLP assays, as well as the identification of the crossover/breakpoint for GYPB DEL2. Using 393 samples from three study sites in Ghana as well as samples from HapMap and 1000 G projects for validation, we show that our assays are sensitive and reliable for genotyping GYPB DEL1 and DEL2. To the best of our knowledge, this is the first report of such high-throughput genotyping assays by PCR-RFLP for identifying specific GYPB deletion types in populations. These assays will enable better identification of GYPB deletions for large genetic association studies and functional experiments to understand the role of this gene cluster region in susceptibility to malaria and other diseases.

Genomic analysis of SARS-CoV-2 reveals local viral evolution in Ghana.

The confirmed case fatality rate for the coronavirus disease 2019 (COVID-19) in Ghana has dropped from a peak of 2% in March to be consistently below 1% since May 2020. Globally, case fatality rates have been linked to the strains/clades of circulating severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) within a specific country. Here we present 46 whole genomes of SARS-CoV-2 circulating in Ghana, from two separate sequencing batches: 15 isolates from the early epidemic (March 12-April 1 2020) and 31 from later time-points ( 25-27 May 2020). Sequencing was carried out on an Illumina MiSeq system following an amplicon-based enrichment for SARS-CoV-2 cDNA. After genome assembly and quality control processes, phylogenetic analysis showed that the first batch of 15 genomes clustered into five clades: 19A, 19B, 20A, 20B, and 20C, whereas the second batch of 31 genomes clustered to only three clades 19B, 20A, and 20B. The imported cases (6/46) mapped to circulating viruses in their countries of origin, namely, India, Hungary, Norway, the United Kingdom, and the United States of America. All genomes mapped to the original Wuhan strain with high similarity (99.5-99.8%). All imported strains mapped to the European superclade A, whereas 5/9 locally infected individuals harbored the B4 clade, from the East Asian superclade B. Ghana appears to have 19B and 20B as the two largest circulating clades based on our sequence analyses. In line with global reports, the D614G linked viruses seem to be predominating. Comparison of Ghanaian SARS-CoV-2 genomes with global genomes indicates that Ghanaian strains have not diverged significantly from circulating strains commonly imported into Africa. The low level of diversity in our genomes may indicate lower levels of transmission, even for D614G viruses, which is consistent with the relatively low levels of infection reported in Ghana.

Analysis of Plasmodium falciparum Rh2b deletion polymorphism across different transmission areas.

Despite significant progress in controlling malaria, the disease remains a global health burden. The intricate interactions the parasite Plasmodium falciparum has with its host allows it to grow and multiply in human erythrocytes. The mechanism by which P. falciparum merozoites invade human erythrocytes is complex, involving merozoite proteins as well as erythrocyte surface proteins. Members of the P. falciparum reticulocyte binding-like protein homolog (PfRh) family of proteins play a pivotal role in merozoite invasion and hence are important targets of immune responses. Domains within the PfRh2b protein have been implicated in its ability to stimulate natural protective antibodies in patients. More specifically, a 0.58 kbp deletion, at the C-terminus has been reported in high frequencies in Senegalese and Southeast Asian parasite populations, suggesting a possible role in immune evasion. We analysed 1218 P. falciparum clinical isolates, and the results show that this deletion is present in Ghanaian parasite populations (48.5% of all isolates), with Kintampo (hyper-endemic, 53.2%), followed by Accra (Hypo-endemic, 50.3%), Cape Coast (meso-endemic, 47.9%) and Sogakope (meso-endemic, 43.15%). Further analysis of parasite genomes stored in the MalariaGEN database revealed that the deletion variant was common across transmission areas globally, with an overall frequency of about 27.1%. Interestingly, some parasite isolates possessed mixed PfRh2b deletion and full-length alleles. We further showed that levels of antibodies to the domain of PfRh2 protein were similar to antibody levels of PfRh5, indicating it is less recognized by the immune system.

Highlights on the Application of Genomics and Bioinformatics in the Fight Against Infectious Diseases: Challenges and Opportunities in Africa.

Genomics and bioinformatics are increasingly contributing to our understanding of infectious diseases caused by bacterial pathogens such as Mycobacterium tuberculosis and parasites such as Plasmodium falciparum. This ranges from investigations of disease outbreaks and pathogenesis, host and pathogen genomic variation, and host immune evasion mechanisms to identification of potential diagnostic markers and vaccine targets. High throughput genomics data generated from pathogens and animal models can be combined with host genomics and patients' health records to give advice on treatment options as well as potential drug and vaccine interactions. However, despite accounting for the highest burden of infectious diseases, Africa has the lowest research output on infectious disease genomics. Here we review the contributions of genomics and bioinformatics to the management of infectious diseases of serious public health concern in Africa including tuberculosis (TB), dengue fever, malaria and filariasis. Furthermore, we discuss how genomics and bioinformatics can be applied to identify drug and vaccine targets. We conclude by identifying challenges to genomics research in Africa and highlighting how these can be overcome where possible.