Predictive Model

Thomas Churcher of Imperial College London in the United Kingdom will develop an analytical method to more accurately determine the origin of new cases of malaria in regions with low levels of the disease, which is critical for elimination efforts. Current methods are unreliable as there are no standardized criteria and they often rely solely on interviews. They will use routinely collected travel history data and disease maps to develop rigorous methods for estimating whether a case of malaria is either acquired locally or imported from another region.

Honorine Ward of Tufts Medical Center in the U.S. will develop a three-dimensional model of the human intestine for rapid screening of drugs targeting the parasite Cryptosporidium, which causes potentially lethal diarrhea in young children in developing countries. Developing drugs against Cryptosporidium has been particularly difficult, partly because of the limited understanding of the parasites behavior in the human intestine, and particularly of the effect of malnutrition, which commonly co-occurs with infection and likely contributes to disease severity.

Samuel Arnold of the University of Washington in the U.S. will develop methods to evaluate drug candidates for treating Cryptosporidium infections, which cause severe diarrhea particularly in young children from developing countries. There are no effective drugs against the Cryptosporidium parasite. This is partly because when it infects humans it becomes isolated in specific cells lining the gastrointestinal tract, which is where a drug would also need to be located at sufficient concentrations to be effective.

Kathryn Colborn of University of Colorado Denver in the U.S. will develop a statistical model to predict future outbreaks of malaria and help identify the most effective intervention strategy. Current models can help work out where and why malaria outbreaks occur rather than predicting future outbreaks. They will use supervised machine learning to develop a set of predictive algorithms using available data including weather, demographics, and malaria incidence in children under five years old from Mozambique.

Xun Suo of China Agricultural University in China will develop a rabbit model of cryptosporidiosis that mimics the human disease, which presents as severe diarrhea particularly in young children, to help identify new treatments. Current animal models of infection by the parasite Cryptosporidium are suboptimal: mice are not naturally infected, while pigs and calves can be infected but are expensive and more difficult to manage, and none show the same symptoms as humans. Rabbits are naturally infected by Cryptosporidium and display human-related symptoms.

Anastasios Tsaousis of the University of Kent in the United Kingdom will build a screening platform to identify drugs that can be used to treat diarrhea caused by the parasite Cryptosporidium, which is the second major cause of death in children under five years old in developing countries. There are currently no effective drugs for treating Cryptosporidium, largely because it cannot easily be grown in the laboratory making it difficult to study and test for new drugs.

Thomas Egwang of Med Biotech Laboratories in Uganda will develop a reformulated medium derived from mammalian cell culture medium that is optimized for storage and preservation of stool-derived helminth eggs. Diagnosis of soil-transmitted helminth infections, which cause considerable morbidity in developing countries, involves microscopic examination of fecal samples. The accuracy of these methods depends on how well the sample has been preserved since collection.

Arash Shaban-Nejad of the University of Tennessee Health Science Center in the U.S. will develop an analytic framework to help integrate dynamic surveillance data from multiple sources and health systems to support decision making for malaria elimination. Data on malaria is currently scattered in different formats across diverse organizations, making it difficult to access and use. An ontology is a web-based method that explicitly defines specific concepts using logical rules and constraints, and can be used to capture and combine information from numerous sources into a formal framework.

Marcos Barreto of Universidade Federal da Bahia in Brazil will build a platform that routinely integrates surveillance data from malaria with socioeconomic and health care data, and also provides open access and support for data analysis and mining. To monitor the spread of malaria in a populous country like Brazil requires an open access surveillance system that can incorporate multiple types of data to support elimination efforts.

Richard Guerrant of the University of Virginia in the U.S. and co-investigators will develop and validate non-invasive metabolic biomarkers of gut health to identify children at risk of environmental enteropathy and developmental impairment, in order to assess interventions. They will use ongoing MAL-ED (malnutrition and enteric diseases) and NIH-supported clinical studies in malnourished and control children, and their own studies in novel murine models, along with a nuclear magnetic resonance approach to perform metabolic profiling of urine, plasma and feces samples.