Electronic Theses and Dissertations (Masters)

Permanent URI for this collection

https://hdl.handle.net/10539/38221

Browse

Now showing 1 - 20 of 53

The role of regulatory T cells in adults in South Africa with active tuberculosis
(2010-01-28T10:11:24Z) Mayne, Elizabeth Sarah
Introduction: Regulatory T cells (Tregs) are increasingly being recognized as key immunological players in immunosuppression and have been seen to be permissive for certain infections. Aim: This study aimed to elucidate the role that Tregs play in symptomatic infection with Mycobacterium tuberculosis (TB), both with and without co-infection with human immunodeficiency virus type 1 (HIV 1) by quantification of these cells at ex vivo. It was then attempted to characterise the behaviour of FoxP3 positive cells in culture with stimulation. Methods: Peripheral blood mononuclear cells were purified from uninfected controls, patients with active TB, patients with HIV infection and patients with HIV infection and active TB. The frequencies of Tregs were assessed by flow cytometry at ex vivo and again after four days of culture with stimulation with anti-CD3, Purified protein derivative, tetanus toxoid and HIV peptide superpools (gag and nef). These frequencies were compared between the four groups of patients. The ability of Tregs and effector T cells to proliferate was also assessed. Interferon-γ secretion was used as a measure of effector T cell response to stimulation. vi Results: Frequencies of Tregs were significantly reduced in patients with active TB as compared with HIV infected patients and uninfected controls. Co-infected individuals showed a broad range of frequencies which were not significantly different from controls. These frequencies remained stable in culture with the exception of those individuals infected with HIV who showed a decline in the frequency of those cells expressing FoxP3 over the period. Cells expressing FoxP3 were not anergic and responded to stimulation. HIV specific proteins, in addition, resulted in specific effects on the Tregs with a positive interferon response to gag correlating with increased Treg frequencies and FoxP3 expression in CD4+ T cells correlated with the proliferative response of CD4+ T cells to Nef in HIV infected individuals. Conclusions: This study shows significant differences of frequencies of FoxP3 positive producing cells in the peripheral blood at ex vivo in patients with active TB. The function of these cells in this population is uncertain and further functional data and long-term clinical follow-up is required. In addition, the frequencies of these cells remained constant over time and showed proliferative response to stimuli (most notably CD3) suggesting that these cells may be generated in the periphery.
Unmasking serial murder: a comparison of a South African murder series with characteristics from the Federal Bureau of Investigation Serial Murder Database
(2015) Holland, Shakeera
The term ‘serial killer’ brings to mind notorious criminals whose crimes are so heinous as to test the limits of the most vivid imagination and make us question their humanity. What is the reality of serial murder? In 2005, the Federal Bureau of Investigation (FBI) hosted a symposium on serial murder, which brought together international experts in the field of serial murder with the aim of clarifying and understanding this multifarious crime. On the 12th of March 2008, Gcinumzi Richman Makhwenkwe, ‘The Moffat Park Serial Murderer’ was convicted of 5 counts of murder, 3 counts of rape and 3 counts of robbery with aggravating circumstances. The Department of Forensic Medicine and Pathology of the University of the Witwatersrand, based at the Johannesburg Forensic Pathology Service (FPS) Medicolegal Mortuary Facility performed the medicolegal investigations of death in all the victims. This research report explores the characteristics of serial murder and serial murderers as documented in the literature; documents the features and characteristics of the Moffat Park murder series; compares the features of this South African murder series to those from the findings of the FBI serial murder symposium; explores the role of the forensic medical practitioner in the investigation of the Moffat Park series and serves to educate and inform forensic medical practitioners of the features of serial murder as awareness may potentially lead to earlier identification of a murder series. This could ultimately lead to earlier implementation of specialist investigative methods, earlier apprehension of the serial murderer and most importantly fewer victims.
Rationalising laboratory workflow to improve the efficiency of diagnostic service delivery: A critical review of haematological malignancies by flow cytometric immunophenotyping at the Charlotte Maxeke Johannesburg Academic Hospital
(University of the Witwatersrand, Johannesburg, 2023) Naidoo, Maynolia; Glencross, Debbie
The 2006 Bethesda medical indications guideline provides concise indications for flow cytometric immunophenotyping (FCI), to enable rationalising a decision for sample processing. The local practice of processing every sample received for FCI places an enormous burden on the resources of the laboratory, and leads to unnecessary expenditure for state health. A ‘triage’ process based on the current Bethesda medical indications guideline may be beneficial in developing countries. The aim of this study was to determine how the implementation of a triage process would impact on the diagnostic service delivery, and the ability to detect or miss disease. A retrospective review of 500 bone marrow aspirate (BMA) samples submitted for FCI analysis was performed from October to December 2019. The sensitivity, specificity, and predictive values of the BMA cytomorphology against the FCI outcomes (‘test-all’) were determined. Thereafter, the Bethesda medical indications guideline was retrospectively applied to the same data set (‘triage’), to compare the decision to process or not to process samples, against objective evidence of disease in various BMA investigations. After exclusion of inadequate quality samples that preclude comparison, 429 ‘test-all’ and 455 triage cases were evaluated. The ‘test-all’ analysis revealed a 97.1% sensitivity and 89.8% specificity, with a 64.1% positive predictive value (PPV) but striking 99.4% negative predictive value (NPV). The triage was largely effective in identifying cases with disease, revealing a 100% sensitivity and 83.3% specificity, with a PPV of 32.5% and very high NPV of 93.8%. Without impacting clinical outcomes, the implementation of a triage process can reduce the burden of FCI testing by 18%. Preliminary cytomorphological review of the accompanying BMA is strongly recommended as an additional step to improve the overall PPV of the triage, while safely reducing unnecessary FCI sample processing in a further 56% of cases. The implementation of a triage process with modifications for local use in flow cytometry laboratories, would enable the appropriate rationalisation of resources, improve the cost- effectiveness, and overall diagnostic service delivery in developing countries like Sub-Saharan Africa
Audit of Lysosomal Storage Diseases Testing at the National Health Laboratory Service in Johannesburg from 2011-2020
(2023) Novellie, Michael
Lysosomal Storage Diseases (LSDs) are a group of Inborn Errors of Metabolism (IEM), due to the lack of a lysosomal enzyme. This results in toxic accumulation of metabolic waste products in various organs leading to neurodevelopmental regression, organ failure and premature death in the absence of treatment. Treatments for LSDs are limited. This study audited LSD diagnostic test requests received by the Division of Human Genetics, National Health Laboratory Service (NHLS) in Johannesburg from 2011 to 2020 with the aim of understanding the demand, appropriateness, and patient management of suspected LSD cases. A quantitative survey of all samples (1861 tests) referred to NHLS Johannesburg during the study period was performed. A total of 198 (13.3%) samples were rejected for testing mainly because of faulty sample collection. Of the 1663 that were accepted for testing 1457 (87.6%) tested negative, 73 (4.4%) were inconclusive and 133 (8.0%) tested positive. Fifty-five (3.1%) patients with LSD test requests, all of which were positive, were known to a Clinical Genetics unit. The most frequently requested test was for Fabry disease: 620 (33.3% of all requests), even though this disease is not the most prevalent LSD. Of the 603 accepted test requests for Fabry disease, only 6 (1.0%) tested positive. This suggests that some referring clinicians had unrealistic expectations of encountering this disease. It should be noted, however, that testing for Fabry disease is part of a broad diagnostic workup that may be applied even if the indication for testing is not specific. Access to LSD testing was unequal: private facilities were proportionally over-represented compared to public facilities; certain provinces with large referral centres (in KZN and Gauteng) were over-represented compared to smaller centres. Feedback and education of referring clinicians regarding indications for testing and importance of patient follow up, especially by clinical genetics services, are recommended. Follow up of positive MPS screening tests with specific diagnostic tests is essential. A system should be implemented where a medical geneticist phones the referring clinician and discusses further sample requirements (blood for enzyme analysis) and referral to a genetics clinic for all positive LSD screening tests. Future consideration should be given to designing a more systematic testing process, with the introduction of molecular testing to supplement biochemical testing.
Laboratory Evaluation of Aspergillus Galactomannan Lateral Flow Assays
(University of the Witwatersrand, Johannesburg, 2023) Ubbink, Anja; Chibabhai, Vindana
Invasive aspergillosis diagnosis is based on a combination of clinical, radiological, and mycological factors, including the detection of Aspergillus galactomannan antigen in serum and bronchoalveolar lavage fluid (BALF). Lateral flow assays (LFA) introduced for rapid detection of galactomannan in serum and BALF include the IMMY sōna Aspergillus LFA (IMMY LFA) and the Dynamiker QuicGMTM Aspergillus Galactomannan Ag LFA (QuicGM LFA). Objective To evaluate the performance of the IMMY LFA and QuicGM LFA in South Africa. Methods Serum and BALF samples previously tested by Platelia BioRad Aspergillus GM-EIA were analysed using the two different LFAs. Percentage agreement and precision was assessed. Results Forty-six serum- and 13 BALF samples were tested using the IMMY GM LFA and 48 serum- and 6 BALF samples were tested using the QuicGM LFA. Using an optical density ≥0.5 as positive, results were compared to the BioRad Aspergillus GM-EIA. For the IMMY LFA, serum samples had a positive percent agreement (PPA) of 0% (0/1); negative percent agreement (NPA) of 91% (41/45) and overall percent agreement (OPA) of 89% (41/46). BALF samples had a PPA of 75% (3/4), NPA 50% (5/10) and OPA of 57% (8/14). For the QuicGM LFA, serum samples had a PPA 0% (0/3), NPA of 96% (43/45) and OPA of 90% (43/48). BALF samples had a PPA of 100% (1/1), NPA of 100% (5/5) and OPA of 100% (6/6). For the IMMY LFA, between-day reproducibility for 72% (13/18) and 63% (5/8) for serum and BALF samples, respectively. Between-batch reproducibility was 89% (16/18) and 50% (4/8), respectively for serum and BALF samples. For the QuicGM LFA, between-day reproducibility was 75% (9/12) and 75% (3/4) for the serum and BALF samples, respectively. The between-batch reproducibility was 100% (8/8) for serum and 100% (3/3) for BALF. Conclusion A follow-up evaluation with a larger sample size utilizing clinical, radiological, and laboratory data is warranted to determine the assays’ clinical utility. What this study adds Invasive aspergillosis is a life-threatening disease, where a prompt diagnosis improves outcome. Currently there is no Aspergillus galactomannan assay available in the South African state sector. This study evaluates two lateral flow assays for the detection of Galactomannan in South Africa
Mutation Profiling of Paediatric Solid Tumours in a Cohort of South African Patients
(University of the Witwatersrand, Johannesburg, 2023) Manolas, Erin; Krause, Amanda; Lamola, Lindie
Childhood cancers are an emerging global health burden, with the highest increase in incidence and mortality rates occurring in low-middle income countries, such as South Africa (SA). Adding to this burden is the contribution of cancer-predisposing genes (CPGs), whose germline variants increase the risk of cancer development in childhood. These genes are largely associated with a set of disorders, known as cancer-predisposing syndromes (CPSs), which are characterised by an increased likelihood of cancer development and/or additional phenotypic malignant/non- malignant features. Next-Generation Sequencing (NGS) technologies have been key in determining the occurrence and contribution of such germline variants to paediatric cancer development across international research and diagnostic settings. However, these technologies have not been applied to paediatric cohorts in SA and thus there is a paucity of data regarding the contribution of germline variants to childhood cancer development in this setting. Through the design and evaluation of an NGS targeted-CPG panel, this study aimed to generate germline variant profiles of SA paediatric cancer patients, thereby gaining insight into their potential role in the pathogenesis of childhood cancers. NGS was performed on the genomic DNA from 32 solid-tumour paediatric cancer patients using an Ion Ampliseq 50 CPG panel design and the Ion Torrent S5 sequencing instruments. Germline variants were called using the Ion Torrent Suite™ software (v.5.12.0) and annotated using the Ensembl Variant Effect Predictor. Variants were filtered using a bioinformatics pipeline assessing variant data from population and public databases, computational data, functional studies data, genetic pedigrees indicating family history and phenotypic data. Variant evidence was further interpreted for variant prioritisation and classification according to the ACMG-AMP guidelines. All putative pathogenic/likely-pathogenic (P/LP) variants identified were validated via Sanger Sequencing. Seven pathogenic and/or likely pathogenic germline variants were identified and validated in seven patients. Three of these variants, identified in the NF1, RET, and TP53 genes, were detected in patients who presented with phenotypes consistent with their genetic findings and are associated with well-known CPSs (diagnostic yield - 3/32, ~9.4%). The remaining four variants, identified in the BRCA1, ERCC3, FAH, and RB1 genes, have not been previously associated with the patient’s cancer phenotype and therefore require further investigation. At the time of this project’s data generation, this is the first global report of the novel heterozygous, likely pathogenic FAH p.R162H variant. Additionally, although reported elsewhere, the majority of the variants identified in this study (6/7, ~86.7%) have been reported for the first time within the SA paediatric population. To our knowledge, this is the first study to utilise NGS technologies in the germline variant profiling of paediatric solid-tumour patients in SA and therefore has greatly added to filling the current knowledge gap. In addition, these findings have contributed towards the foundation for the development of a CPG sequencing panel suitable for implementation in a SA diagnostic setting
Evaluation of novel assay formats for indoleamine 2, 3 dioxygenase as a tb biomarker
(University of the Witwatersrand, Johannesburg, 2023) Tsolo, Semakaleng Theresia; Ranchod, Heena
The World Health Organization has prioritized the development of non-sputum-based assays that are capable of detecting active Tuberculosis (TB). Tryptophan (tryp) is converted to kynurenine (kyn) by the rate-limiting enzyme indoleamine 2, 3- dioxygenase (IDO). IDO activity may serve as a biomarker for active TB. Dried blood spots (DBS) can be collected outside of medical institutions and are simple to transport. We wanted to explore the use of DBS as an alternative sample type to measure the kyn/tryp ratio and IDO mRNA gene expression in healthy people. METHODS We optimised methods for elution of dried blood spots, exploring various elution buffers. Following method optimisation, we enrolled 40 healthy participants, and collected whole blood and DBS samples. Kyn and tryp concentrations were measured using ELISA (ImmuSmol, France). IDO mRNA gene expression was determined by real-time PCR using two housekeeping genes GAPDH and BACT. Statistical analysis was performed to determine the correlation and agreement between peripheral blood samples and DBS. RESULTS For IDO activity, tryp showed good agreement between plasma and DBS with a median percentage similarity of 91.1%. In contrast, no agreement was observed for kyn with a median percentage similarity of 56.6%. The kyn/tryp ratio performed poorly due to poor detection of kyn in DBS. Percentage similarity between whole blood and DBS IDO mRNA expression using GAPDH 87.1%, while using BACT was 84.6%. Using either traditional sample types or DBS, there was no correlation between IDO gene expression and kyn/tryp ratio. CONCLUSION We showed that tryp was measurable in DBS. Tryp in DBS was 91.1% similar to values in plasma. Despite method optimization, there was poor agreement between DBS and plasma for kyn. Although IDO mRNA gene expression was detectable in DBS, method agreement with whole blood was unsatisfactory. Alternative methods for the stabilization of kyn in DBS should be explored in future studies. IDO mRNA expression should be measured from whole blood in future studies
A review of genetic testing for females referred to genetic counselling for a personal or family history of gynaecological cancer
(University of the Witwatersrand, Johannesburg, 2023) Barnard, Sebastian
Hereditary cancer syndromes, caused by pathogenic variants in specific genes, substantially increase an individual’s risk for cancer and are estimated to cause 10% of all uterine cancers and 20% of all ovarian cancers. However, these data are primarily based on high-income countries and to date there is no published data on the known variants and testing of cancer predisposition genes associated with gynaecological cancers in South Africa. In this study, patient records for those with either a confirmed diagnosis or family history of gynaecological cancer that were seen by the Division of Human Genetics (NHLS/Wits) were retrospectively analysed (n = 104). Associations between patient characteristics, genetic testing availability and the detection of pathogenic variants as well as the utility of risk assessment tools were investigated using statistical analysis. The majority of patients underwent diagnostic genetic testing (78/104, 75.0%), 25 (32.1%) were positive, 41 (52.6%) were negative, and 12 (15.4%) returned a variant of unknown significance. Test results were significantly different between European and non-European patients (p << 0.05) with non-European patients being 30% less likely to have a pathogenic variant detected (OR 0.7, 95% CI 0 0.22, 2.21). Patients who met genetic testing criteria according to online risk assessments were more likely to have a positive genetic test result than those who did not (p < 0.05). A disparity exists not only in genetic testing availability but also clinic attendance between public and private healthcare which is likely limiting the ability to diagnose hereditary cancer syndromes associated with gynaecological cancers in public healthcare hospitals
Exome Sequencing of individuals with vitiligo and their relatives with and without autoimmune disorders
(University of the Witwatersrand, Johannesburg, 2023) Rabinda, Kentie Rofhiwa
Autoimmune disorders result from the immune system attacking and damaging its healthy cells because of an acquired immune system malfunction. Vitiligo is an example of an autoimmune disorders. It is a common depigmentation skin disorder with an estimated prevalence of 0.5–2% of the population worldwide. It is shown by selective loss of melanocytes causing non- scaly, chalky-white macules. Individuals with vitiligo often have other autoimmune disorders or first-degree relatives with at least one other autoimmune disorder. This study aimed to identify genetic variants in selected genes in individuals with vitiligo and their unaffected close relatives who have other autoimmune disorders. Upon DNA extraction, whole exome sequencing was performed, and five non- major histocompatibility complex (MHC) genes were analysed. Identified variants were annotated on Ensembl Variant Effect Predictor (VEP) and compared between individuals with vitiligo and their close relatives with and without autoimmune disorders in a family-specific manner. Eight variants were identified in two families, however, the three missense variants in the NLRP1 gene (rs12150220, rs11651270, and rs2301582) were observed between and across two families in individuals with autoimmune disorders. SMOC2 rs13208776, was identified as a risk locus for vitiligo in individuals from two Romanian isolate villages with vitiligo and other autoimmune disorders. None of the genotyped individuals were homozygous for the minor allele (A) and 2/12 individuals were heterozygous therefore it is unlikely that rs13208776 has any role in the development of vitiligo or any of the autoimmune disorders present in the two families. This study showed that NLRP1 gene variants segregated with vitiligo and autoimmune disorders
Analysis of Whole Exome Sequence Data from African Patients with HD-Like Features and No Known HDPhenocopy Syndrome
(University of the Witwatersrand, Johannesburg, 2023) Naicker, Racilya; Krause, Amanda; Baine-Savanhu, Fiona
Huntington disease (HD) is a rare progressive neurodegenerative disorder that results from a CAG repeat expansion within the huntingtin gene (HTT). Several syndromes present with HD- like features in the absence of the HTT expansion and are termed HD phenocopies. Huntington disease-like 2 (HDL2), a known phenocopy, is most commonly observed in individuals with African, or probable African, ancestry. Therefore, previous diagnostic testing in the Division of Human Genetics, National Health Laboratory Service (Johannesburg, South Africa) screened for both HD and HDL2 in patients with HD-like phenotypes and an indication of African ancestry. Patients who tested negative for both syndromes remain undiagnosed, highlighting the need for further testing strategies. This study aimed to identify variants implicated in the HD-like phenotype of these patients. In a group of nine undiagnosed patients with Black African ancestry, a virtual gene panel was analysed through a whole exome sequencing (WES) approach. The data was filtered, and candidate variants were prioritised according to the frequency, type, and location of the variants as well as in-silico pathogenicity prediction scores. A total of 20 candidate variants in 15 genes were shortlisted and classified according to ACMG/AMP guidelines. Notably, variants in the mitochondrial DNA polymerase subunit gamma (POLG; c.2246T>C; p.Phe749Ser) and the glutaryl-CoA dehydrogenase (GCDH; c.877G>A; p.Ala293Thr) genes were classified as likely pathogenic and pathogenic, respectively. Genotype-phenotype correlation indicated a potential diagnosis of autosomal dominant progressive external ophthalmoplegia in the patient carrying the POLG variant, whereas the GCDH variant was considered an incidental finding due to a lack of correlation with the characteristics of glutaric aciduria type 1. These findings highlight the diagnostic challenges faced in the African context for patients with HD-like clinical features and call for further validation studies and re-analysis of the WES data using alternative gene panels for the patients for whom no putative pathogenic variants were identified
Sequencing for high-risk type 1 diabetes genotypes in the South African black population using AmpliSeq Nanopore next generation sequencing
(University of the Witwatersrand, Johannesburg, 2023) Mathabela, Nomfundo Mathabela; Ali, Stuart
Introduction: Type 1 diabetes (T1D) is a chronic autoimmune disorder characterized by the destruction of the -cells of the pancreas, resulting in the inability to produce/maintain insulin. This results in an inability to maintain the blood glucose at homeostasis. Prolonged hyperglycaemia leads to micro- and macro-vascular complications. Thus, it is vital to diagnose and treat patients in a timely manner. It is important to identify individuals at increased risk of developing T1D allowing for appropriate follow ups. Numerous mutations/variants in specific genes confer an increased risk of T1D with the HLA gene accounting for approximately 40-50 % of the risk. Therefore, it is possible that by looking at genetic variation in T1D associated genes we can develop a tool that can determine the likelihood of an individual developing T1D. Study aims and objectives: The development, implementation, and validation of a Nanopore NGS method to sequence and genotype polymorphisms associated with T1D in a cohort of black South Africans which could be used to develop a genetic risk score (GRS) for T1D in this population. In addition, we aimed to compare sequencing results to two HLA genotypes obtained using PCR-RFLP. Method: Participants with T1D (n=19) and control participants (n=5) were genotyped for 12 T1D associated polymorphisms through Ampliseq Nanopore sequencing. In addition, Ion Torrent was used to confirm the Ampliseq Nanopore results. A bioinformatics pipeline that involves reference sequence generation using an in-house script, alignment of sequencing data with the reference sequence, filtering and variant calling was developed. A genetic risk score (GRS) was calculated for the participants. Participants (n=73) were genotyped by PCR-RFLP for the HLA rs2040410 and rs7454108 polymorphisms. Results: Sequencing samples individually was found to have a slightly higher Qscore than samples sequenced in multiplex (9.73 vs. 9.5). Samples sequenced individually had higher average reads (187.60 vs. 151.14 Mb), passed reads (41.47 vs. 25.99 Mb), and estimated bases (54.72 vs. 49.57 Mb) than those sequenced in multiplex. In addition, samples sequenced in a multiplex had higher average failed reads (475.58 Mb) in comparison to those sequenced individually (13.58 Mb). The average percentage difference in sequencing data generated using Ion Torrent compared to Nanopore was 5.67%. Variant calling produced average Phred-scale quality scores of 73.89 for standards (The Coriell Trio) and 89.77 for participants. The GRS calculator was not able to accurately predict which participants had T1D. Discussion and Conclusion: A next generation sequencing method and bioinformatics pipeline for the screening of participants for 645 T1D associated polymorphisms was investigated. The method combined two sequencing techniques i.e., Ion AmpliSeq and Nanopore sequencing to achieve this. The data can then be processed by in-house variant callers. With a larger sample group, this method will be useful for the investigation of genetic variants linked to T1D.
Germline Cancer Predisposition Variants in Paediatric Rhabdomyosarcoma
(University of the Witwatersrand, Johannesburg, 2023) Pillhofer, Gabriella Peta; Lamola, Lindie; Mnika, Khuthala
Rhabdomyosarcoma (RMS) is cancer that originates from undifferentiated skeletal muscle cells. RMS is the most common tissue sarcoma in children and adolescents and has over 50% of cases occurring in individuals under the age of 10 and thus there is reason to suspect that RMS may have a hereditary predisposition. However, much of the existing research of germline predisposition in paediatric RMS is focused on ethnically European populations, and currently very little data is available from African populations. In this study, we aimed to identify RMS predisposition variants by performing whole exome sequencing (WES) on germline DNA from eight paediatric patients diagnosed with RMS. Following WES, variant annotation and filtering was performed to identify variants in genes of interest that were potentially germline causes of malignancy. Filtered variants were then classified according to American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG-AMP) guidelines. This study identified four variants of uncertain significance (VUSs) in four patients. Two of these were in genes that have previously been associated with an RMS phenotype (SDHB and BRCA1), and two were in genes that have been associated with a hereditary cancer syndrome that is not linked to RMS (CBL and CREBBP). While the highlighted variants are not of clinical significance, this study emphasises the importance of cataloguing and reporting VUSs in research in Africa. By expanding the genomic database on African patients, the analysis of variants on the continent may be made more accurate and efficient. It is the goal that the knowledge gained from this study will contribute to the information base of hereditary paediatric cancers in Africa, and that it may encourage similar research so that the field may continue to expand.
Machine Learning on biochemical data for the prediction of mutation presence in suspected Familial Hypercholesterolaemia
(2024) Hesse, Reinhardt
Background Familial hypercholesterolemia (FH) is a common monogenic disorder and, if not diagnosed and treated early, results in premature atherosclerotic cardiovascular disease. Most individuals with FH are undiagnosed due to limitations in current screening and diagnostic approaches, but the advent of machine learning (ML) offers a new prospect to identify these individuals. Our objective was to create a ML model from basic lipid profile data with better screening performance than low-density lipoprotein cholesterol (LDL-C) cut-off levels and diagnostic performance comparable to the Dutch Lipid Clinic Network (DLCN) criteria. Methods The ML model was developed using a combination of logistic regression, deep learning and random forest classification and was trained on a 70% split of an internal dataset consisting of 555 individuals clinically suspected of having FH. The performance of the model, as well as that of the LDL-C cut-off and DLCN criteria, were assessed on both the internal 30% testing dataset and a high prevalence external dataset by comparing the area under the receiver operator characteristic (AUROC) curves. All three methodologies were measured against the gold standard of FH diagnosis by mutation identification. Furthermore, the ML model was also tested on two lower prevalence datasets derived from the same external dataset. Results The ML model achieved an AUROC curve of 0.711 on the high prevalence external dataset (n=1376; FH prevalence=64%), which was superior to that of the LDL-C cut off alone (AUROC=0.642) and comparable to that of the DLCN criteria (AUROC=0.705). The model performed even better when tested on the medium prevalence (n=2655; FH prevalence=20%) and low prevalence (n=1616; FH prevalence=1%) datasets, with AUROC curve values of 0.801 and 0.856 respectively. Conclusions Despite the absence of clinical information, the ML model was better at correctly identifying genetically confirmed FH in a cohort of individuals suspected of having FH than the LDL-C cut-off tool and comparable to the DLCN criteria. The same ML model performed even better when tested on two cohorts with lower FH prevalence. The application of ML is therefore a promising tool in both the screening for, and diagnosis of, individuals with FH.
Molecular epidemiology of M and E protein coding genes from South African SARS-CoV-2 strains, 2020 to 2021
(2024) Marsden, Fabian
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the current pathogen causing the COVID-19 pandemic across the world. While vaccines that elicit anti-SARS-CoV2 antibodies have been developed and licenced, there is a reduced protection against variants of concern (VOCs) such as Beta, Delta and Omicron. This is due to the mutations within the spike (S) protein which is the antigen targeted by most vaccines. Other potential vaccine targets include the structural proteins namely the membrane (M) and envelope (E) proteins of SARSCoV-2 which are more conserved. In this study we aimed to determine the extent of genetic diversity in the M and E protein genes from South African SARS-CoV-2 strains and its impact on predicted B and T cell epitopes. M and E gene sequences were extracted from South African SARS-CoV-2 genomes obtained from the Global Initiative on Sharing All Influenza Data (GISAID) database for the period 01March 2020 to 31 December 2021. Maximum-likelihood phylogenetic tree analysis shows that among South African E gene sequences only the Omicron VOC sequences form a distinct cluster. Similarly, the Omicron M gene sequences also form a distinct cluster compared to the Wuhan reference strain, Beta and Delta sequences. The predicted T cell and B cell epitopes of M and E proteins were identified with specific regions that have shown to have identical regions in both the variants and the reference strain, this shows the conserved nature of the M and E genes. SARS-CoV-2 are shown to have varying antigenic probabilities for M and E proteins from each of the variants considered as probable antigens. The allergenicity and toxicity of the M and E proteins was assessed in the context of potential vaccine development with certain peptides of each shown to have toxic properties. The predicted B and T cell epitopes show that despite the presence of mutations in the VOCs’ derived protein sequences, there is a common epitopic region that is shared between the reference and the variants. There is a strong 9-mer coverage by the natural sequences despite some non-coverage due to non-silent mutations. The results from the epitope predication and HLA typing shows the conserved nature of the M and E proteins which highlights the potential use for the development of vaccines.
Maternal vaccination in South Africa: timing and completeness
(2024) Bourne, Julia
Maternal immunisation is an invaluable public health measure that protects not only the mother, but also the foetus and new-born infant against a host of diseases; and is recommended by both the World Health Organisation (WHO) and South African national health authorities. Pregnancy induces a heightened state of immune system vulnerability, leaving women more susceptible to severe influenza outcomes, whilst neonatal tetanus has a fatality rate of between 80-100% in the absence of medical intervention. Maternal immunisation against influenza and tetanus has been successfully utilised as a public health strategy across the globe to uphold maternal and neonatal health. Maintaining coverage is imperative for both diseases as influenza strains change seasonally and tetanus cannot be eliminated, highlighting the importance of continued maternal immunisation. This study aimed to describe the uptake of both influenza and tetanus vaccinations during pregnancy, the completion of the tetanus vaccination schedule and the timing of both influenza and tetanus immunisation within South African antenatal care facilities. In addition, this study described clinical and demographic factors affecting maternal immunisation uptake. Clinical and demographic data were collected in a parent study and were retrospectively analysed in this study using the statistical software Stata. Influenza vaccination uptake within the sampled population was found to be 16.62% (806/4851). The odds of influenza vaccination were significantly higher in women aged 21-30 years, and women with six or more ANC visits. Metro East Cape Town site in the Western Cape outperformed Gauteng sites, with significantly increased odds of influenza vaccination amongst women frequenting that site. Appropriate influenza immunisation: defined as immunisation occurring during the period of either 01/04/2017-31/07/2017 or 01/04/2018-30/06/2018, occurred in 74.86% (530/708) of the cohort. Women who were alcohol users were significantly more likely to receive an influenza vaccine – yet this may be explained by the Metro East site which had the higher influenza coverage having the highest prevalence of alcohol use during pregnancy. Of 7105 women, 7031 (98.96%) received at least one dose of tetanus toxoid vaccine (TTV). Of these women, 39.24% (2759) received one dose; 51.06% (3590) received two doses and 9.70% (682) received the recommended three doses of TTV in their index pregnancy. Tetanus schedule completion was significantly more likely in women ≤20 years, and those who presented for their booking antenatal care (ANC) visit in the first trimester. In addition, women with more than three visits had an increased likelihood of TTV schedule completion. The odds of TTV schedule completion were decreased by negatively parity and gravidity, values over one and less than six, and greater than one respectively. Women with hypertension were significantly less likely to receive three TTV doses compared to women without hypertension. Julia Bourne MSc(Vaccinology) Research Report: Version 2.0 v Tetanus immunisation schedule adherence prevalence was 0.60% (4/670) in women with three doses and 90.34% (3209/3552) in women with two. Improvements may be made in South African maternal immunisation coverage, with this study supporting the idea of targeted educational campaigns and a revision of the maternal immunisation schedule to include the tetanus, diphtheria & acellular pertussis vaccine instead of the tetanus toxoid vaccine.
The use of insecticide treated eave ribbons as a protection tool against pyrethroid-resistant populations of mosquitoes that transmit malaria and dengue fever
(2024) Shirima, Ruth Severin
Vector control methods such as insecticide treated nets (ITNs) and indoor residual spraying (IRS) have been successful in preventing mosquito-borne diseases like malaria and dengue. However, these methods face challenges including insecticide resistance, high costs, logistical difficulties, low adoption rates, and limited durability. Therefore, there is a need for simpler and more affordable interventions that can be used on a large scale in disease-endemic communities to supplement current approaches. This study evaluated the effectiveness of using insecticide-treated eave ribbons as a potential tool for complementing the current vector control methods. Eave ribbons are pieces of hessian fabric that can be placed around the eave spaces of poorly constructed houses to kill or repel mosquitoes. Laboratory cone bioassays were conducted to assess the efficacy of eave ribbons treated with the organophosphate, pirimiphos-methyl, for killing the malaria vectors, Anopheles funestus and Anopheles arabiensis, and the dengue vector, Aedes aegypti, under varying exposure durations and insecticide doses. In addition, a semi-field experiment was done to assess the efficacy of eave ribbons treated with pirimiphosmethyl against the malaria vectors. Indoor and outdoor biting was assessed by the number of mosquitoes captured indoors in window exit traps and outdoors by human landing catches, respectively. Mortality of recaptured mosquitoes was recorded after 24, 48, and 72 hours. The findings revealed that treated eave ribbons resulted in higher mosquito mortality than the untreated ribbons, but the impact increased with increased exposure duration or dose. The semi-field study indicated moderate levels of bite prevention and mortality of the mosquitoes. At the doses of 1 g a.i./m2 and 2 g a.i./m2 pirimiphos-methyl, there was no significant protection against An. arabiensis, but at the dose of 4 g a.i./m2 pirimiphos-methyl, there was only significant protection against outdoor biting An. arabiensis, but not An. funestus. In conclusion, while insecticide-treated eave ribbons may have potential for controlling malaria and dengue vectors, further research is needed to validate their efficacy in field settings and to identify suitable insecticides or insecticide combinations that are safe and effective.
Impact of donor CYP3A5 genotype on pharmacokinetics of tacrolimus in South African paediatric liver transplant patients
(2024) Wheeler, Caitlin
Tacrolimus is characterised by a narrow therapeutic target range and wide interpatient variability. Pharmacogenetic research attributes CYP3A5 single nucleotide polymorphisms to the variable inter-patient tacrolimus concentration dose ratios (CDR). In this study, we compared the mean tacrolimus CDR in paediatric liver transplant recipients and their living liver donors’ CYP3A5 rs776746 T>C genotypes(*1/*1, *1/*3 and *3/*3), accounting for donor and recipient characteristics. The graft-to-recipient weight ratio and the CYP3A5 donor genotypes were found to be independent factors significantly impacting the mean tacrolimus CDR. Donor CYP3A5 expressors (*1/*1 and *1/*3) were shown to have significantly lower recipient tacrolimus CDRs, therefore, higher dosages would be required in comparison to CYP3A5 non-expressors to reach the same therapeutic target range. The potential implementation of a stratified medicine dosage algorithm, combining living liver donor CYP3A5 genotyping with the calculation of the graft-to-recipient weight ratio, may predict the optimal tacrolimus dosage schedules for liver transplant recipients.
Characterisation of RSV fusion proteins from South African patients with RSV disease, 2019 to 2020
(2024) Mabilo, Prince
Respiratory syncytial virus (RSV) is classified into subtypes A (RSV-A) and B (RSV-B), which are classified into different genotypes based on genetic variability of the G surface glycoprotein gene. The F surface protein gene is more conserved however variability in signal peptide, transmembrane domain, and antigenic sites have been reported. The study was conducted in the Virology laboratory, Charlotte Maxeke Johannesburg Academic Hospital (CMJAH), South Africa. Study participants included patients of all ages from whom respiratory samples were submitted for respiratory viruses diagnoses from 2019 to 2020. The complete RSV F genes were amplified, library prep was preformed using the Nextera DNA prep kits and sequenced using amplicon-based next generation sequencing on the Illumina MiSeq sequencing platform. The Genome Detective Virus tool v2.27 was used to assemble sequencing reads and MEGA X was used for phylogenetic analysis. N-linked glycosylation and amino acid sequence variation was assessed. The overall prevalence of RSV was 5.8% (101/1 734). Seventy (69.3%; 70/101) RSV-positive samples were available for genetic characterisation of the F protein gene and thirty-one (30.7%; 31/101) were excluded due to insufficient sample volume. Only RSV-A strains were identified (91.2%; 31/34). Twenty three of thirty-one (74.2%) of the RSV F gene sequences from 2019 to 2020 clustered together with bootstrap values ranging from 64% to 99% and were NA1-like. A N-glycosylation site at position 120 gained by South African strains from 2018 is retained in strains from this study. This N-glycosylation site is present in approximately 25.8% of RSV strains from this study. The diversity of RSV-A F proteins was low, with amino acid variations observed at 30/571 (5.3%) sites. Ten mutations were detected in 4/6 antigenic sites (I, II, IV and V), with frequencies ranging from 0.3 to 100%. Antigenic changes seen exclusively among South African strains are: Y33H (0.3%) and V384T (7.3%) at site I and S275F (0.3%) at site II. Seven mutations associated with escape of human leukocyte antigen (HLA)-restricted cytotoxic T lymphocyte (CTL) were predicted in seven epitopes. Overall amino acid mutation frequency for 2019 to 2020 RSV-A F genes is similar to that reported for South African strains from 2018 (3.3% to 6.7%). For the first time in South Africa we detected the S275F mutation which causes palivizumab resistance.
Exploring the interaction of host genetics and the gut microbiome in obesity in an African population
(2024) Schnell, Samantha Susan
Obesity is a highly prevalent health concern that is on the rise in Sub-Saharan Africa. Even though genetic variation and gut microbiota have been implicated in the development of obesity independently, the interactions between these factors have not been previously explored in a South African cohort. This study aimed to identify possible associations between host genomes, body mass index as a measure of obesity, and gut microbiota composition (in the form of V3-V4 16S rRNA sequencing) in a female African cohort. Polygenic risk scores predictive of body mass index in this cohort were generated to categorise the participants into high- and low-risk groups. Subsequently, several statistical analyses were performed comparing gut microbiota between these groups. High-risk participants with high body mass index had associations identified with increased abundances of Prevotella_9 and VadinBE97. In contrast, the low polygenic risk and low body mass index sub-group was associated with greater Bacteroides levels. This study acknowledges the plausible interactions between these factors in an African cohort.
Evaluating the clinical utility of a SNP-based microarray platform: a comparative study
(2024) Mayisela, Minenhle Peniel
Developmental disorders make up a majority of cases seen in genetic clinics at the National Health Laboratory Service (NHLS). Chromosomal microarray analysis (CMA) is the first line testing for individuals with developmental disorders and congenital anomalies. This study compared the clinical utility of single nucleotide polymorphism (SNP) - based array (CytoScan® Optima array) and a comparative genomic hybridization (CGH) array (SurePrint G3 Unrestricted CGH ISCA v2, 8x60 kb microarray) in diagnosis of developmental disorders. This was done by looking at the differences in copy number variant calls, segment size differences and gene content between the two microarrays. The cost to run each type of microarray test for diagnosis was also compared. The copy number variant calls made by the two platforms were comparable. The SNP-based array did provide additional information that would be useful in molecular diagnosis. The final recommendation was for the CGH array and the SNP-based array to be used interchangeably based on clinical requests at the Division of Human Genetics

Browse

Browsing Electronic Theses and Dissertations (Masters) by Issue Date

Results Per Page

Sort Options