Draft Genome Sequence of an Enterotoxigenic Escherichia coli Strain Carrying Genes for Colonization Surface Antigen 13 and a Heat-Labile Toxin.

Here, we report the draft genome of ESEI_597, an enterotoxigenic Escherichia coli (ETEC) strain harboring genes encoding colonization surface antigen 13 (CS13) and a heat-labile toxin. The ESEI_597 strain was isolated from an 8-month-old child living in Korogocho, Kenya, in 2013.

Using Machine Learning Methods Incorporating Individual Reader Annotations to Classify Paediatric Chest Radiographs in Epidemiological Studies.

Introduction: Epidemiological studies that involve interpretation of chest radiographs (CXRs) suffer from inter-reader and intra-reader variability. Inter-reader and intra-reader variability hinder comparison of results from different studies or centres, which negatively affects efforts to track the burden of chest diseases or evaluate the efficacy of interventions such as vaccines. This study explores machine learning models that could standardize interpretation of CXR across studies and the utility of incorporating individual reader annotations when training models using CXR data sets annotated by multiple readers. Methods: Convolutional neural networks were used to classify CXRs from seven low to middle-income countries into five categories according to the World Health Organization's standardized methodology for interpreting paediatric CXRs. We compared models trained to predict the final/aggregate classification with models trained to predict how each reader would classify an image and then aggregate predictions for all readers using unweighted mean. Results: Incorporating individual reader's annotations during model training improved classification accuracy by 3.4% (multi-class accuracy 61% vs 59%). Model accuracy was higher for children above 12 months of age (68% vs 58%). The accuracy of the models in different countries ranged between 45% and 71%. Conclusions: Machine learning models can annotate CXRs in epidemiological studies reducing inter-reader and intra-reader variability. In addition, incorporating individual reader annotations can improve the performance of machine learning models trained using CXRs annotated by multiple readers.

Using self-supervised feature learning to improve the use of pulse oximeter signals to predict paediatric hospitalization.

Background: The success of many machine learning applications depends on knowledge about the relationship between the input data and the task of interest (output), hindering the application of machine learning to novel tasks. End-to-end deep learning, which does not require intermediate feature engineering, has been recommended to overcome this challenge but end-to-end deep learning models require large labelled training data sets often unavailable in many medical applications. In this study, we trained self-supervised learning (SSL) models for automatic feature extraction from raw photoplethysmography (PPG) obtained using a pulse oximeter, with the aim of predicting paediatric hospitalization.  Methods: We compared logistic regression models fitted using features extracted using SSL with models trained using both clinical and SSL features. In addition, we compared end-to-end deep learning models initialized randomly or using weights from the SSL models. We also compared the performance of SSL models trained on labelled data alone (n=1,031) with SSL trained using both labelled and unlabelled signals (n=7,578). Results: Logistic regression models were more predictive of hospitalization when trained on features extracted using labelled PPG signals only compared to SSL models trained on both labelled and unlabelled signals (AUC 0.83 vs 0.80). However, features extracted using SSL model trained on both labelled and unlabelled PPG signals were more predictive of hospitalization when concatenated with clinical features (AUC 0.89 vs 0.87). The end-to-end deep learning model had an AUC of 0.80 when initialized using the SSL model trained on all PPG signals, 0.77 when initialized using SSL trained on labelled data only, and 0.73 when initialized randomly. Conclusions: This study shows that SSL can extract features from PPG signals that are predictive of hospitalization or initialize end-to-end deep learning models. Furthermore, SSL can leverage larger unlabelled data sets to improve performance of models fitted using small labelled data sets.

Levels of PM10 and PM2.5 and Respiratory Health Impacts on School-Going Children in Kenya.

BACKGROUND: The respiratory system of children is vulnerable to exposure to particulate matter (PM) with a diameter of less than 2.5 and 10 µm (PM2.5 and PM10) or even lower. OBJECTIVE: This study assessed PM10 and PM2.5 levels and respiratory health impacts on children in schools located in an industrialized suburb in Kenya. METHOD: The PM10 and PM2.5 levels were sampled from five public primary schools in Athi River Township and a control school during the wet and dry seasons. Outdoor and classroom samples were collected concurrently on an 8-hour mean during school hours on two consecutive days in each school and analyzed using gravimetric techniques. Five hundred and seventy-eight (n = 578) pupils aged 9-14 years from these schools were also evaluated for symptoms of respiratory illnesses and lung function using a questionnaire and spirometric method, respectively, during the same periods. RESULTS: Indoor median PM10 levels (µg/m3) ranged from 60.8-269.1 and 52.8-232.3 and PM2.5 values (µg/m3) of 17.7-52.4 and 28.5-75.5 during the dry and wet seasons, respectively. The control classrooms had significantly (p <0.05) lower median PM10 levels (µg/m3) of 5.2 and 4.2, and PM2.5 levels (µg/m3) of 3.5 and 3.0 during the respective seasons. Nearly all the classrooms in Athi River schools had PM2.5 and PM10 median levels that exceeded the World Health Organization (WHO) recommended levels. The indoor-to-outdoor ratios varied from 0.35-1.40 and 0.80-2.40 for PM10 and 0.30-0.80 and 0.80-1.40 for PM2.5 during the dry and wet seasons, respectively, suggesting higher levels in the classrooms during the wet season. The relative risk (RR) and odds ratio (OR) presented higher prevalence of respiratory diseases following PM exposure in all the Athi River schools than the control during the dry and wet seasons. At 95% CI, the RR and OR showed strong associations between high PM10 and PM2.5 levels and lung function deficits and vice versa. The association was more prevalent during the wet season. CONCLUSIONS: The study calls for effective indoor air management programs in school environments to reduce PM exposure and respiratory health impacts. PARTICIPANT CONSENT: Obtained. ETHICS APPROVAL: The research permit and approvals were obtained from the University of Nairobi/Kenyatta National Hospital Ethics and Research Committee (KNH-UoN ERC Reference: P599/08/2016) and the National Commission for Science, Technology and Innovation (Reference: NACOSTI/P/18/4268/25724). COMPETING INTERESTS: The authors declare no competing financial interests.

Exploration of miRNA families for hypotheses generation.

Technological improvements have resulted in increased discovery of new microRNAs (miRNAs) and refinement and enrichment of existing miRNA families. miRNA families are important because they suggest a common sequence or structure configuration in sets of genes that hint to a shared function. Exploratory tools to enhance investigation of characteristics of miRNA families and the functions of family-specific miRNA genes are lacking. We have developed, miRNAVISA, a user-friendly web-based tool that allows customized interrogation and comparisons of miRNA families for hypotheses generation, and comparison of per-species chromosomal distribution of miRNA genes in different families. This study illustrates hypothesis generation using miRNAVISA in seven species. Our results unveil a subclass of miRNAs that may be regulated by genomic imprinting, and also suggest that some miRNA families may be species-specific, as well as chromosome- and/or strand-specific.