Background Methicillin resistant Staphylococcus aureus (MRSA) strains were once confined to hospitals however, in the last 20 years MRSA infections have emerged in the community in people with no prior exposure to hospitals. Strains causing such infections were novel and referred to as community-associated MRSA (CA-MRSA). The aim of this study was to determine the MRSA carriage rate in children in eastern Uganda, and to investigate coexistence between CA-MRSA and hospital-associated (HA-MRSA). Methods Between February and October 2011, nasopharyngeal samples (one per child) from 742 healthy children under 5 years in rural eastern Uganda were processed for isolation of MRSA, which was identified based on inhibition zone diameter of ≤19 mm on 30 μg cefoxitin disk. SCC mec and spa typing were performed for MRSA isolates. Results A total of 140 S. aureus isolates (18.9%, 140/742) were recovered from the children of which 5.7% (42/742) were MRSA. Almost all (95.2%, 40/42) MRSA isolates were multidrug resistant (MDR). The most prevalent SCC mec elements were types IV (40.5%, 17/42) and I (38.1%, 16/42). The overall frequency of SCC mec types IV and V combined, hence CA-MRSA, was 50% (21/42). Likewise, the overall frequency of SCC mec types I, II and III combined, hence HA-MRSA, was 50% (21/42). Spa types t002, t037, t064, t4353 and t12939 were detected and the most frequent were t064 (19%, 8/42) and t037 (12%, 5/42). Conclusion The MRSA carriage rate in children in eastern Uganda is high (5.7%) and comparable to estimates for Mulago Hospital in Kampala city. Importantly, HA-MRSA (mainly of spa type t037) and CA-MRSA (mainly of spa type t064) coexist in children in the community in eastern Uganda, and due to high proportion of MDR detected, outpatient treatment of MRSA infection in eastern Uganda might be difficult. Electronic supplementary material The online version of this article (10.1186/s13756-019-0551-1) contains supplementary material, which is available to authorized users.

Background : Genetic studies of biomedical phenotypes in underrepresented populations identify disproportionate numbers of novel associations. However, current genomics infrastructure--including most genotyping arrays and sequenced reference panels--best serves populations of European descent. A critical step for facilitating genetic studies in underrepresented populations is to ensure that genetic technologies accurately capture variation in all populations. Here, we quantify the accuracy of low-coverage sequencing in diverse African populations. Results : We sequenced the whole genomes of 91 individuals to high-coverage ( > 20X) from the Neuropsychiatric Genetics of African Population-Psychosis (NeuroGAP-Psychosis) study, in which participants were recruited from Ethiopia, Kenya, South Africa, and Uganda. We empirically tested two data generation strategies, GWAS arrays versus low-coverage sequencing, by calculating the concordance of imputed variants from these technologies with those from deep whole genome sequencing data. We show that low-coverage sequencing at a depth of ≥4X captures variants of all frequencies more accurately than all commonly used GWAS arrays investigated and at a comparable cost. Lower depths of sequencing (0.5-1X) performed comparable to commonly used low-density GWAS arrays. Low-coverage sequencing is also sensitive to novel variation, with 4X sequencing detecting 45% of singletons and 95% of common variants identified in high-coverage African whole genomes. Conclusion : These results indicate that low-coverage sequencing approaches surmount the problems induced by the ascertainment of common genotyping arrays, including those that capture variation most common in Europeans and Africans. Low-coverage sequencing effectively identifies novel variation (particularly in underrepresented populations), and presents opportunities to enhance variant discovery at a similar cost to traditional approaches.

Background : Genetic studies of biomedical phenotypes in underrepresented populations identify disproportionate numbers of novel associations. However, current genomics infrastructure--including most genotyping arrays and sequenced reference panels--best serves populations of European descent. A critical step for facilitating genetic studies in underrepresented populations is to ensure that genetic technologies accurately capture variation in all populations. Here, we quantify the accuracy of low-coverage sequencing in diverse African populations. Results : We sequenced the whole genomes of 91 individuals to high-coverage ( > 20X) from the Neuropsychiatric Genetics of African Population-Psychosis (NeuroGAP-Psychosis) study, in which participants were recruited from Ethiopia, Kenya, South Africa, and Uganda. We empirically tested two data generation strategies, GWAS arrays versus low-coverage sequencing, by calculating the concordance of imputed variants from these technologies with those from deep whole genome sequencing data. We show that low-coverage sequencing at a depth of ≥4X captures variants of all frequencies more accurately than all commonly used GWAS arrays investigated and at a comparable cost. Lower depths of sequencing (0.5-1X) performed comparable to commonly used low-density GWAS arrays. Low-coverage sequencing is also sensitive to novel variation, with 4X sequencing detecting 45% of singletons and 95% of common variants identified in high-coverage African whole genomes. Conclusion : These results indicate that low-coverage sequencing approaches surmount the problems induced by the ascertainment of common genotyping arrays, including those that capture variation most common in Europeans and Africans. Low-coverage sequencing effectively identifies novel variation (particularly in underrepresented populations), and presents opportunities to enhance variant discovery at a similar cost to traditional approaches.

During a period of eight months, the carcases of 16,800 slaughter cattle were inspected at a city abattoir in Uganda. Eighty-seven of them had tuberculosis-like lesions and tissue samples were cultured. Only 17 cultures yielded acid-fast bacilli; 11 of them were confirmed as Mycobacterium bovis and six as non-tuberculous mycobacteria (NTM). GenoType Mycobacterium assays on the six NTM identified two as Mycobacterium fortuitum and one as Mycobacterium intracellulare, but three were unidentified. Characterisation of the M bovis isolates by spoligotyping and IS6110 restriction fragment length polymorphism (RFLP) revealed that five of the six spoligotype patterns observed in the 11 strains had not been previously reported, and seven of the nine isolates typed by RFLP had multicopy number IS6110 patterns. Six of the 11 infected carcases had multiple sites of infection, but none was condemned as unfit for human consumption.

In Uganda, Artemether-Lumefantrine and Artesunate are recommended for uncomplicated and severe malaria respectively, but are currently threatened by parasite resistance. Genetic and epigenetic factors play a role in predisposing Plasmodium falciparum parasites to acquiring Pfkelch13 (K13) mutations associated with delayed artemisinin parasite clearance as reported in Southeast Asia. In this study, we report on the prevalence of mutations in the K13, pfmdr-2 (P. falciparum multidrug resistance protein 2), fd (ferredoxin), pfcrt (P. falciparum chloroquine resistance transporter), and arps10 (apicoplast ribosomal protein S10) genes in Plasmodium falciparum parasites prior to (2005) and after (2013) introduction of artemisinin combination therapies for malaria treatment in Uganda. A total of 200 P. falciparum parasite DNA samples were screened. Parasite DNA was extracted using QIAamp DNA mini kit (Qiagen, GmbH, Germany) procedure. The PCR products were sequenced using Sanger dideoxy sequencing method. Of the 200 P. falciparum DNA samples screened, sequencing for mutations in K13, pfmdr-2, fd, pfcrt, arps10 genes was successful in 142, 186, 141, 128 and 74 samples respectively. Overall, we detected six (4.2%, 6/142; 95%CI: 1.4–7.0) K13 single nucleotide polymorphisms (SNPs), of which 3.9% (2/51), 4.4% (4/91) occurred in 2005 and 2013 samples respectively. All four K13 SNPs in 2013 samples were non-synonymous (A578S, E596V, S600C and E643K) while of the two SNPs in 2005 samples, one (Y588N) is non-synonymous and the other (I587I) is synonymous. There was no statistically significant difference in the prevalence of K13 (p = 0.112) SNPs in the samples collected in 2005 and 2013. The overall prevalence of SNPs in pfmdr-2 gene was 39.8% (74/186, 95%CI: 25.1–50.4). Of this, 4.2% (4/95), 76.9% (70/91) occurred in 2005 and 2013 samples respectively. In 2005 samples only one SNP, Y423F (4.2%, 4/95) was found while in 2013, Y423F (38.5%, 35/91) and I492V (38.5%, 35/91) SNPs in the pfmdr-2 gene were found. There was a statistically significant difference in the prevalence of pfmdr-2 SNPs in the samples collected in 2005 and 2013 (p<0.001). The overall prevalence of arps10 mutations was 2.7% (2/72, 95%CI: 0.3–4.2). Two mutations, V127M (4.5%: 1/22) and D128H (4.5%: 1/22) in the arps10 gene were each found in P. falciparum parasite samples collected in 2013. There was no statistically significant difference in the prevalence of arps10 SNPs in the samples collected in 2005 and 2013 (p = 0.238). There were more pfmdr-2 SNPs in P. falciparum parasites collected after introduction of Artemisinin combination therapies in malaria treatment. This is an indicator of the need for continuous surveillance to monitor emergence of molecular markers of artemisinin resistance and its potential drivers in malaria affected regions globally.