Use of unsupervised machine learning to characterise HIV predictors in sub-Saharan Africa.

INTRODUCTION: Significant regional variations in the HIV epidemic hurt effective common interventions in sub-Saharan Africa. It is crucial to analyze HIV positivity distributions within clusters and assess the homogeneity of countries. We aim at identifying clusters of countries based on socio-behavioural predictors of HIV for screening. METHOD: We used an agglomerative hierarchical, unsupervised machine learning, approach for clustering to analyse data for 146,733 male and 155,622 female respondents from 13 sub-Saharan African countries with 20 and 26 features, respectively, using Population-based HIV Impact Assessment (PHIA) data from the survey years 2015-2019. We employed agglomerative hierarchical clustering and optimal silhouette index criterion to identify clusters of countries based on the similarity of socio-behavioural characteristics. We analyse the distribution of HIV positivity with socio-behavioural predictors of HIV within each cluster. RESULTS: Two principal components were obtained, with the first describing 62.3% and 70.1% and the second explaining 18.3% and 20.6% variance of the total socio-behavioural variation in females and males, respectively. Two clusters per sex were identified, and the most predictor features in both sexes were: relationship with family head, enrolled in school, circumcision status for males, delayed pregnancy, work for payment in last 12 months, Urban area indicator, known HIV status and delayed pregnancy. The HIV positivity distribution with these variables was significant within each cluster. CONCLUSIONS /FINDINGS: The findings provide a potential use of unsupervised machine learning approaches for substantially identifying clustered countries based on the underlying socio-behavioural characteristics.

Use of machine learning techniques to identify HIV predictors for screening in sub-Saharan Africa.

AIM: HIV prevention measures in sub-Saharan Africa are still short of attaining the UNAIDS 90-90-90 fast track targets set in 2014. Identifying predictors for HIV status may facilitate targeted screening interventions that improve health care. We aimed at identifying HIV predictors as well as predicting persons at high risk of the infection. METHOD: We applied machine learning approaches for building models using population-based HIV Impact Assessment (PHIA) data for 41,939 male and 45,105 female respondents with 30 and 40 variables respectively from four countries in sub-Saharan countries. We trained and validated the algorithms on 80% of the data and tested on the remaining 20% where we rotated around the left-out country. An algorithm with the best mean f1 score was retained and trained on the most predictive variables. We used the model to identify people living with HIV and individuals with a higher likelihood of contracting the disease. RESULTS: Application of XGBoost algorithm appeared to significantly improve identification of HIV positivity over the other five algorithms by f1 scoring mean of 90% and 92% for males and females respectively. Amongst the eight most predictor features in both sexes were: age, relationship with family head, the highest level of education, highest grade at that school level, work for payment, avoiding pregnancy, age at the first experience of sex, and wealth quintile. Model performance using these variables increased significantly compared to having all the variables included. We identified five males and 19 females individuals that would require testing to find one HIV positive individual. We also predicted that 4·14% of males and 10.81% of females are at high risk of infection. CONCLUSION: Our findings provide a potential use of the XGBoost algorithm with socio-behavioural-driven data at substantially identifying HIV predictors and predicting individuals at high risk of infection for targeted screening.

Antiplasmodial prenylated flavanonols from Tephrosia subtriflora.

The CH2Cl2/MeOH (1:1) extract of the aerial parts of Tephrosia subtriflora afforded a new flavanonol, named subtriflavanonol (1), along with the known flavanone spinoflavanone B, and the known flavanonols MS-II (2) and mundulinol. The structures were elucidated by the use of NMR spectroscopy and mass spectrometry. The absolute configuration of the flavanonols was determined based on quantum chemical ECD calculations. In the antiplasmodial assay, compound 2 showed the highest activity against chloroquine-sensitive Plasmodium falciparum reference clones (D6 and 3D7), artemisinin-sensitive isolate (F32-TEM) as well as field isolate (KSM 009) with IC50 values 1.4-4.6 µM without significant cytotoxicity against Vero and HEp2 cell lines (IC50 > 100 µM). The new compound (1) showed weak antiplasmodial activity, IC50 12.5-24.2 µM, but also showed selective anticancer activity against HEp2 cell line (CC50 16.9 µM).

Risk of Fungi Associated with Aflatoxin and Fumonisin in Medicinal Herbal Products in the Kenyan Market.

Utilization of herbal products is a major concern due to the possibility of contamination by toxigenic fungi that are mycotoxin producers such as Aspergillus species during processing and packaging. Research was carried out to determine the presence of aflatoxins and fumonisins in herbal medicinal products sold in Eldoret and Mombasa towns in Kenya. The study employed both exploratory and laboratory experimental design. The herbal products were purchased from the market and transported to Kenya Medical Research Institute for processing and analysis. Fungal contaminants were determined according to Pharmacopoeia specifications. The toxins were quantified using ELISA based technique. The genus Aspergillus was the most dominant followed by Penicillium. Fungal counts ranged between 1 CFU/g and >1000 cfu/g. Analysis of variance showed that the rate of fungal contaminants for Eldoret and Mombasa samples had significant association (p ≤ 0.001). Aflatoxin levels ranged from 1 to 24 ppb, while fumonisin levels ranged from 1 to >20 ppb. Only 31% of samples met the standards for microbial limits as specified in Pharmacopoeia. There is need for product microbial quality improvement through proper harvesting, processing, storage, and marketing. It is recommended that a policy be enacted to enable regulation of herbal products in Kenya.

High-throughput screening and sensitized bacteria identify an M. tuberculosis dihydrofolate reductase inhibitor with whole cell activity.

Mycobacterium tuberculosis (Mtb), the causative agent of tuberculosis, is a bacterial pathogen that claims roughly 1.4 million lives every year. Current drug regimens are inefficient at clearing infection, requiring at least 6 months of chemotherapy, and resistance to existing agents is rising. There is an urgent need for new drugs that are more effective and faster acting. The folate pathway has been successfully targeted in other pathogens and diseases, but has not yielded a lead drug against tuberculosis. We developed a high-throughput screening assay against Mtb dihydrofolate reductase (DHFR), a critical enzyme in the folate pathway, and screened a library consisting of 32,000 synthetic and natural product-derived compounds. One potent inhibitor containing a quinazoline ring was identified. This compound was active against the wild-type laboratory strain H37Rv (MIC(99) = 207 µM). In addition, an Mtb strain with artificially lowered DHFR levels showed increased sensitivity to this compound (MIC(99) = 70.7 µM), supporting that the inhibition was target-specific. Our results demonstrate the potential to identify Mtb DHFR inhibitors with activity against whole cells, and indicate the power of using a recombinant strain of Mtb expressing lower levels of DHFR to facilitate the discovery of antimycobacterial agents. With these new tools, we highlight the folate pathway as a potential target for new drugs to combat the tuberculosis epidemic.

The potential of the extracts of Tagetes minuta Linnaeus (Asteraceae), Acalypha fruticosa Forssk (Euphorbiaceae) and Tarchonanthus camphoratus L. (Compositae) against Phlebotomus duboscqi Neveu Lemaire (Diptera: Psychodidae), the vector for Leishmania major Yakimoff and Schokhor.

BACKGROUND & OBJECTIVES: Harmful effects of synthetic chemical insecticides including vector resistance, environmental pollution and health hazards have necessitated the current significance in the search for plant-based insecticide products that are environmentally safe and effective to leishmaniases control. The insecticidal activity of Tagetes minuta Linnaeus (Asteraceae), Acalypha fruticosa Forssk (Euphorbiaceae) and Tarchonanthus camphoratus L. (Compositae) extracts were investigated against Phlebotomus duboscqi Neveu Lemaire (Diptera: Psychodidae). METHODS: The extracts were prepared from dried aerial parts soaked in methanol and ethyl acetate twice until the filtrates became clear, filtered and dried out by rotary evaporation at 30-35 degrees C. The solid extracts obtained were later prepared into 2.5, 5 and 10 mg/ml. Two millilitres of the solutions were blotted on filter papers, which were dried overnight and placed into jars where adult sandflies were aspirated. Males and females were assayed separately. RESULTS & CONCLUSION: The extracts had significant mortality (p<0.05) in both males and females bioassays but were not significantly different between sexes. The extracts of Acalypha fruticosa and Tagetes minuta had significantly higher mortality rates than those of Tarchonanthus camphoratus and the different concentrations used showed significantly different mortality rates and 10 mg/ml was the most effective concentration. Cent percent mortality was obtained at 96 h of exposure to 5 and 10 mg/ml concentrations except for Tarchonanthus camphoratus which had a mortality of only 46.7% in 10 mg/ml bioassay. These extracts were found to be insecticidal to adult sandflies.

In vitro effects of Warburgia ugandensis, Psiadia punctulata and Chasmanthera dependens on Leishmania major promastigotes.

Plant extracts from Warburgia ugandensis Sprague (Family: Canellaceae), Psiadia punctulata Vatke (Family: Compositae) and Chasmanthera dependens Hoschst (Family: Menispermaceae) were tested for activity on Leishmania major promastigotes (Strain IDU/KE/83 = NLB-144) and infected macrophages in vitro. Plants were collected from Baringo district, dried, extracted, weighed and tested for antileishmanial activity. Serial dilutions of the crude extracts were assayed for their activity against Leishmania major in cell free cultures and in infected macrophages in vitro. Inhibitory concentrations and levels of cytotoxicity were determined. Warburgia ugandensis, Psiadia punctulata and Chasmanthera dependens had an IC(50) of 1.114 mg/ml, 2.216 mg/ml and 4.648 mg/ml, respectively. The cytotoxicity of the drugs on BALB/c peritoneal macrophage cells was insignificant as compared to the highly toxic drug of choice Pentostam(®). The supernatants from control and Leishmania infected macrophages were analyzed for their nitrite contents by Griess reaction and nitrite absorbance measured at 540 nm. Warburgia ugandensis (stem bark water extract), Chasmanthera dependens (stem bark water extract) and Psiadia punctulata (stem bark methanol extract) produced 112.3%, 94% and 88.5% more nitric oxide than the untreated infected macrophages respectively. Plant crude extracts had significant (p<0.05) anti-leishmanial and immunomodulative effects but insignificant cytotoxic effects at 1mg/ml concentration. All experiments were performed in triplicate. Statistical analysis of the differences between mean values obtained from the experimental group compared to the controls was done by students't test. ANOVA was used to determine the differences between the various treatment groups. The analysis program Probit was used to determine IC(50)s.

Antimicrobial activity of Acacia mellifera extracts and lupane triterpenes.

ETHNOPHARMACOLOGICAL RELEVANCE: Acacia mellifera (Vahl) Benth (Leguminosae) is a subtropical medicinal plant that is widely used in traditional African medicines against various diseases such as pneumonia and malaria. AIM OF THE STUDY: The present study was performed to evaluate the antimicrobial effects of extracts from the stem bark of Acacia mellifera. MATERIALS AND METHODS: The extracts were examined for antibacterial and antifungal activity using the disc-diffusion method against bacterial strains: Streptococcus pneumoniae (ATCC 25923), Pseudomonas aeruginosa (ATCC 27853), Escherichia coli (ATCC 35218), Escherichia coli (ATCC 25922) and Staphylococcus aureus (ATCC 25923) and fungal strains Candida albicans (ATCC 90028), Candida krusei (ATCC 6255), Cryptococcus neoformans (ATCC 6603) and clinical isolates of Microsporum gypseum and Trichophyton mentagrophytes. RESULTS: Some of these extracts were found to be active against some bacterial and fungal strains and were further fractionated to give 12 pure compounds. The methanolic and methanol:dichloromethane (1:1) extracts exhibited antibacterial and antifungal activity. The two were active against Staphylococcus aureus (ATCC 25923), Microsporum gypseum, and Trichophyton mentagrophytes. Activity guided fractionation led to isolation of two active compounds: 3-(Z)-cis coumaroylbetulin and 30-hydroxylup-20 (29)-en-3beta-ol which were against Staphylococcus aureus (ATCC 25923), Microsporum gypseum, Trichophyton mentagrophytes and Pseudomonas aeruginosa (ATCC 27853). CONCLUSIONS: These results may partly explain and support the use of Acacia mellifera stem barks for the treatment of infectious diseases in traditional Kenya medicine.

An ethnobotanical study of medicinal plants used by the Nandi people in Kenya.

ETHNOPHARMACOLOGICAL RELEVANCE: The study of local knowledge about natural resources is becoming increasingly important in defining strategies and actions for conservation or recuperation of residual forests. AIMS OF THE STUDY: This study therefore sought to collect information from local populations concerning the use of Nandi Forest medicinal plants; verify the sources of medicinal plants used and determine the relative importance of the species surveyed. MATERIALS AND METHODS: Data was obtained using semi-structured forms to record the interviewee's personal information and topics related to the medicinal use of specific plants. A total of 40 medicinal plants used locally for the treatment and/or control of human ailments were collected through interviews conducted with selected traditional doctors and professional healers. RESULTS: This study demonstrated that local people tend to agree with each other in terms of the plants use and that leaf material form the major component of plant parts exploited. The other harvested materials consist of stem bark, the roots and the whole plant, though at a lower intensity for making liquid concoctions from different plants. Majority of the remedies were prepared from a single species. In most cases, the mode of administration was oral. In the forest, some of the plants collected were scarce. This scarcity was attributed to indiscriminate logging, overexploitation, poor harvesting methods and current agricultural trends. CONCLUSION: Conservation procedures and creation of awareness were identified as the main remedies to the current situation.

Lupane triterpenoids from Acacia mellifera with cytotoxic activity.

Three new pentacyclic triterpenoids: (20R)-3-oxolupan-30-al (1), (20S)-3-oxolupan-30-al (2) and (20R)-28-hydroxylupen-30-al-3-one (3), along with (20S)-3beta-hydroxylupan-30-al (4), the latter previously described as a constituent of an epimeric mixture, were isolated from Acacia mellifera. In addition, the known metabolites 30-hydroxylup-20-(29)-en-3-one (5), 30-hydroxylup-20-(29)-en-3beta-ol (6), atranorin, methyl 2,4-dihydroxy-3,6 dimethyl benzoate, sitosterol-3beta-O-glucoside and linoleic acid were found in the analyzed plant species for the first time. The structures of the new metabolites were elucidated by extensive spectroscopic analyses and their relative stereochemistry was determined by NOESY experiments. The new metabolite 3 exhibited significant cytotoxic activity against the NSCLC-N6 cell line, derived from a human non-small-cell bronchopulmonary carcinoma.

In vivo screening of antimalarial activity of Acacia mellifera (Benth) (Leguminosae) on Plasmodium berghei in mice.

The activities of total extracts and lupane triterpenes, isolated from the stem bark of Acacia mellifera, were evaluated against Plasmodium berghei strain ANKA in a female Swiss mouse model. Five isolated compounds and the crude extracts were evaluated for antimalarial activity and Quinine hydrochloride was used as a positive control. Only betulin and the methanolic extract produced considerable antimalarial activity in mice infected with P. berghei parasites. This study demonstrated the presence of bioactive agents in Acacia mellifera.

Cytotoxic lupane-type triterpenoids from Acacia mellifera.

One new and eight previously described lupane-type metabolites were isolated for the first time from Acacia mellifera (Leguminosae). Based on spectral analyses, the structure of the new compound was elucidated as 28-hydroxy-3-oxo-lup-20-(29)-en-30-al (1), while the known compounds were identified as 3-oxo-lup-20-(29)-en-30-al (2), 3-hydroxy-lup-20-(29)-en-30-al (3), 28-hydroxy-lup-20-(29)-en-3-one (4), lupenone (5), lupeol (6), betulin (7), betulinic acid (8), and betulonic acid (9). Metabolites 2, 3, and 4 are reported for the first time in the Leguminosae family. The cytotoxicity of the isolated metabolites was evaluated on the NSCLC-N6 cell line, derived from a human non-small-cell bronchopulmonary carcinoma. Compounds 1 and 3 exhibited significant levels of activity.