Biology subjects

Sikulu-Lord, M. T.

Publications and source records attributed to Sikulu-Lord, M. T..

3 recordsLinked to original sources

bioRxiv2023

Identification of visible and near-infrared signature peaks for arboviruses and Plasmodium

Arbovirus and malaria infections affect more than half of the worlds population causing major financial and physical burden. Current diagnostic tools such as microscopy, molecular and serological techniques are technically demanding, costly, or time consuming. Near-infrared spectroscopy has recently been demonstrated as a potential diagnostic tool for malaria and arbovirus and as a screening tool for disease vectors. However, pathogen specific infrared peaks that allow detection of these infections are yet to be described. In this study, we identified unique NIRS peaks from existing laboratory strains of four major arboviruses including Barmah Forest virus (BFV), Dengue virus (DENV), Ross River virus (RRV), Sindbis virus (SINV) and Plasmodium falciparum. Secondly, to determine the diagnostic ability of these peaks, we developed machine learning algorithms using Artificial Neural network (ANN) to differentiate arboviruses from media in which they are grown. Signature peaks for BFV were identified within the visible region at 410, 430, 562 and 588nm and the NIR region at, 946, 958, 1130, 1154 and 1780 nm. DENV related peaks were seen at 410nm within the visible region and 1130 nm within the NIR region. Signature peaks for RRV were observed within the visible region at 410 and 430 nm and within the NIR region at 1130 and 1780 nm, while SINV had a prominent peak at 410 nm within the visible region. Peaks at 514, 528, 547, 561, 582, and 595nm and peaks at 1388, 1432, 1681, 1700, 1721, 1882, 1905, 2245, 2278, 2300 nm were unique for P. falciparum. NIRS predictive sensitivity defined as the ability to predict an arbovirus as an infection was 90% (n = 20) for BFV, 100% (n =10) for RRV and 97.5% (n= 40) for DENV, while infection specificity defined as the ability to predict media as not-infected was 100% (n= 10). Our findings indicate that spectral signatures obtained by NIRS are potential biomarkers for diagnosis arboviruses and malaria. Author summaryMore than half of the world is at risk of contracting either malaria or arboviral infections. In resource limited settings, timely detection of these infections and the ability to screen thousands of people in a day using affordable tools is key to preventing their spread and unprecedented outbreaks. This emphasizes the need to develop portable, rapid, and easy to use diagnostic and surveillance tools. The near-infrared spectroscopy technique has recently been shown to be a potential tool for the diagnosis and surveillance of both malaria and arboviral infections. However, signature spectral biomarkers for these infections remain to be described. This study has identified several spectral biomarkers for DEN, RRV, BVF, SIN arboviruses and P. falciparum parasites. These biomarkers will assist the future assessment of this technique as a diagnostic and or surveillance tool for these infections in the field.

microbiology↗

bioRxiv2020

An Autoencoder and Artificial Neural Network-based Method to Estimate Parity Status of Wild Mosquitoes from Near-infrared Spectra

BackgroundAfter mating, female mosquitoes need animal blood to develop their eggs. In the process of acquiring blood, they may acquire pathogens, which may cause different diseases to humans such as malaria, zika, dengue, and chikungunya. Therefore, knowing the parity status of mosquitoes is useful in control and evaluation of infectious diseases transmitted by mosquitoes, where parous mosquitoes are assumed to be potentially infectious. Ovary dissections, which currently are used to determine the parity status of mosquitoes, are very tedious and limited to very few experts. An alternative to ovary dissections is near-infrared spectroscopy (NIRS), which can estimate the age in days and the infectious state of laboratory and semi-field reared mosquitoes with accuracies between 80 and 99%. No study has tested the accuracy of NIRS for estimating the parity status of wild mosquitoes. Methods and resultsIn this study, we train artificial neural network (ANN) models on NIR spectra to estimate the parity status of wild mosquitoes. We use four different datasets: An. arabiensis collected from Minepa, Tanzania (Minepa-ARA); An. gambiae collected from Muleba, Tanzania (Muleba-GA); An. gambiae collected from Burkina Faso (Burkina-GA); and An.gambiae from Muleba and Burkina Faso combined (Muleba-Burkina-GA). We train ANN models on datasets with spectra preprocessed according to previous protocols. We then use autoencoders to reduce the spectra feature dimensions from 1851 to 10 and re-train ANN models. Before the autoencoder was applied, ANN models estimated parity status of mosquitoes in Minepa-ARA, Muleba-GA, Burkina-GA and Muleba-Burkina-GA with out-of-sample accuracies of 81.9 {+/-} 2.8% (N=927), 68.7 {+/-} 4.8% (N=140), 80.3 {+/-} 2.0% (N=158), and 75.7 {+/-} 2.5% (N=298), respectively. With the autoencoder, ANN models tested on out-of-sample data achieved 97.1 {+/-} 2.2%, (N=927), 89.8 {+/-} 1.7% (N=140), 93.3 {+/-} 1.2% (N=158), and 92.7 {+/-} 1.8% (N=298) accuracies for Minepa-ARA, Muleba-GA, Burkina-GA, and Muleba-Burkina-GA, respectively. ConclusionThese results show that a combination of an autoencoder and an ANN trained on NIR spectra to estimate parity status of wild mosquitoes yields models that can be used as an alternative tool to estimate parity status of wild mosquitoes, especially since NIRS is a high-throughput, reagent-free, and simple-to-use technique compared to ovary dissections.

bioinformatics↗

bioRxiv2018

Age Grading An. Gambiae and An. Arabiensis Using Near Infrared Spectra and Artificial Neural Networks

BackgroundNear infrared spectroscopy (NIRS) is currently complementing techniques to age-grade mosquitoes. NIRS classifies lab-reared and semi-field raised mosquitoes into < or [≥] 7 days old with an average accuracy of 80%, achieved by training a regression model using partial least squares (PLS) and interpreted as a binary classifier.\n\nMethods and findingsWe explore whether using an artificial neural network (ANN) analysis instead of PLS regression improves the current accuracy of NIRS models for age-grading malaria transmitting mosquitoes. We also explore if directly training a binary classifier instead of training a regression model and interpreting it as a binary classifier improves the accuracy.\n\nA total of 786 and 870 NIR spectra collected from laboratory reared An. gambiae and An. arabiensis, respectively, were used and pre-processed according to previously published protocols. Based on ten-fold Monte Carlo cross-validation, an ANN regression model scored root mean squared error (RMSE) of 1.6 {+/-} 0.2 for An. gambiae and 2.8 {+/-} 0.2 for An. arabiensis; whereas the PLS regression model scored RMSE of 3.7 {+/-} 0.2 for An. gambiae, and 4.5 {+/-} 0.1 for An. arabiensis. When we interpreted regression models as binary classifiers, the accuracy of the ANN regression model was 93.7 {+/-} 1.0 % for An. gambiae, and 90.2 {+/-} 1.7 % for An. arabiensis; while PLS regression model scored the accuracy of 83.9 {+/-} 2.3% for An. gambiae, and 80.3 {+/-} 2.1% for An. arabiensis. We also find that a directly trained binary classifier yields higher age estimation accuracy than a regression model interpreted as a binary classifier. A directly trained ANN binary classifier scored an accuracy of 99.4 {+/-} 1.0 for An. gambiae, and 99.0 {+/-} 0.6% for An. arabiensis; while a directly trained PLS binary classifier scored 93.6 {+/-} 1.2% for An. gambiae, and 88.7 {+/-} 1.1% for An. arabiensis.\n\nConclusionTraining both regression and binary classification age models using ANNs yields models with higher estimation accuracies than when the same age models are trained using PLS. Regardless of the model architecture, directly trained binary classifiers score higher accuracy on classifying age of mosquitoes than a regression model translated as binary classifier. Therefore, we recommend training models to estimate age of An. gambiae and An. arabiensis using ANN model architectures and direct training of binary classifier instead of training a regression model and interpret it as a binary classifier.

bioinformatics↗