Waterborne Parasite Infection Risk Mitigation via Microscopy-based Assays and Parasiticide Proteases

Overview

Objectives Waterborne disease remains an economic and health burden to approximately 2 billion people, particularly in impoverished regions. Current control methods for waterborne parasites are highly inefficacious and fail to manage parasite populations prior to infection. The objective of this study is to 1) create an automated assay for parasites in water bodies and 2) computationally develop a class of parasiticide proteases to reduce parasite populations. Methods Adaptive noise inclusion and feature extraction were performed on an image and video dataset of 5 parasite classes (Cryptosporidium, Schistosoma, etc). Individual organisms were then segmented out via Teh-Chain Approximation contouring and compiled into an ordered dataset. To address phenotypic similarity amongst parasites, a Long Short-Term Memory (LSTM) model was employed in conjunction with a Convolutional Neural Network (CNN) for contextualized image classification of the segmented organisms. Next, proteins from WBPS12 and EuPathDB were evaluated via BLAST and a literature search to find target biomarker protease substrates (TBPS) that were expressed externally in each parasite. A script was written to read PeptideCutter and PROSPER tool results for each TBPS and generate the most probable cleavage sites for protease development. PepComposer results were generated for each TBPS and analyzed to finalize the parasiticide protease scaffolds. Results The preliminary segmentation algorithm had an error rate of 1.2% and gathered novel, real-time metrics that were unattainable through previous methodology. Moreover, the hybrid LSTM-CNN model was 97% accurate for multi-class contextualized image classification. Additionally, parasiticide protease scaffolds were evaluated through FoldX energy scores, and 5 were selected with the highest stability and specificity. Conclusions This proposed system consists of the assay (1) to determine which parasites are present, after which the appropriate parasiticide (2) is disseminated to reduce parasite presence in a water body. The automated assay demonstrates unparalleled accuracy and range in parasitological study, and the designed parasiticides target parasites precisely. Furthermore, this system is extensible to other organisms due to the retraining potential of the developed pipeline. Thus, water bodies can be efficiently monitored and treated, thereby mitigating exposure and infection.

Competition history

CSEF 2019 Computational Systems & Analysis (Senior Division Only) · Entry S0817

Resources

CSEF project summary

Related projects

CSEF · 2017

Computationally Designing Antibodies for Target Proteins to Create a Novel Test to Detect Schistosomes in Water

ISEF · 2018

An Affordable, Autonomous, AI-Enhanced Microscope to Enable Efficient Diagnosis of Parasitic Infection in Developing Countries

ISEF · 2017

The Development of a Mechanized Approach to Rapidly and Sensitively Detect and Purify Water Contaminated with Shigella, E. coli, Salmonella, and Cholera through the Use of Carbon-Based Biosensors in Conjunction with Arduino-Controlled Micropipettes

CSEF · 2018

Using Generative Adversarial Networks to Enhance an Affordable Microscope for Epidemic Prevention in Developing Countrie

ISEF · 2021

Detecting Microbial Contaminants in Water Using Artificial Intelligence and Deep Learning Algorithms in order to Prevent the Spread of Waterborne Illnesses

ISEF · 2025

Development and Evaluation of a Novel Smart Automated System for Chromogenic Bacterial Detection in Water Samples

ISEF · 2024

Global Microplastic Pollution - The Emerging Health Crisis: The Ecotoxicological Effects of Microplastics on Aquatic Organisms and Translating to Humans via A.I. Machine Learning & a Novel Low-Cost Water Filtration System

ISEF · 2023

MicroScan: A Computer Vision Tool To Assist Malaria Microscopy Diagnosis

Closest projects by meaning, across every fair and year in the corpus.