Biology subjects

Baldini, F.

Publications and source records attributed to Baldini, F..

At least 19 recordsLinked to original sources

bioRxiv2026

First nationwide mosquito surveillance in Northern Irish wetlands reveals presence of Culex pipiens/molestus hybrids

While mosquito invasions and associated pathogens are increasingly prominent in Europe, evidence gaps remain for many northerly regions where proactive management interventions may be most effective. At the regional periphery of Europe, there remains a significant knowledge gap concerning mosquito vector distributions in Northern Ireland. This study aimed to address this major gap by systematically surveying larval and adult mosquito populations in key habitats (ponds, reedbeds, saltmarsh, wet grasslands, wet woodlands, and urban gardens). We further identified environmental drivers of important vector species abundance and screened collected specimens with molecular techniques to identify presence of cryptic human-biting vector species. In Northern Ireland, mosquitoes (including potential bridge vectors of West Nile and Usutu viruses) were found to be widely distributed. Between May and August 2025, a total of 2430 mosquitoes (951 adults and 1479 larvae) were collected over 196 trap nights and 76 larval collections. Collections encompassed 17 species, with Culex pipiens/torrentium being the most widely distributed and abundant taxon recorded. Mosquito abundances were greatest in reedbed habitats, related negatively with rainfall and peaked in June and July. A subset of Culex pipiens/torrentium mosquitoes were identified to species/biotype level using PCR, with the majority (89.3%) identified as Culex pipiens pipiens and 9.3% as Culex torrentium. Importantly, the first detections of Culex pipiens/molestus hybrids (1.8%) on the island of Ireland were found. This work underscores the wide, yet previously undocumented abundance and distribution of potential zoonotic vectors in Northern Ireland and provides an important baseline for management planning and future research, while contributing to broader understandings of environmental population drivers and distributions.

ecology↗

bioRxiv2026

How reliable is the use of ovarian dilatations for mosquito age-grading? A blinded multi-rater validation of the Polovodova method in Anopheles mosquitoes.

BackgroundAccurate estimation of mosquito age is critical for malaria surveillance, as only older female Anopheles mosquitoes can transmit the parasite. Traditional age-grading methods, include the Polovodova technique in which dilatations on the ovarian pedicel to tally completed gonotrophic cycles (GC) are counted. Despite their widespread use, these techniques remain poorly validated under realistic operational conditions, with limited evidence on their precision and accuracy across different parity rates, GC, and species. In this study, the accuracy and precision of the Polovodova method was evaluated across multiple age-graders and species under both laboratory and field conditions. MethodsA blinded validation study was conducted using laboratory-reared Anopheles arabiensis and Anopheles funestus (N > 3,600) spanning four gonotrophic cycles, and field-collected mosquitoes from Ulanga district, Tanzania (N = 600). Three blinded researchers performed dissections and readings, while a fourth unblinded researcher served as the reference reader (R0) to obtain ground truth (for parity) or assumed truth (for GC) for laboratory mosquitoes. Accuracy and precision for parity and gonotrophic cycle classification were assessed using pairwise inter-rater comparisons, percent agreement, Fleiss kappa and Cohens kappa O_SCPCAP(C_SCPCAPO_SCPLOWKC_SCPLOWO_SCPCAP)C_SCPCAP statistics. For field mosquitoes, which had no age-referenced group, only precision was assessed. FindingsThe Polovodova method accurately classified parity in both species, correctly identifying 98-100% of nulliparous and 94-96% of parous mosquitoes compared with the reference dataset. Inter-rater reliability for parity was almost perfect in An. arabiensis and An. funestus (k = 0.81-0.83), with 82-96% pairwise agreement. GC classification was highly accurate for early stages (GC0-GC2: 85-100%) but declined at later stages GC4(34-46%), reflecting systematic underestimation of mosquito age. Overall GC reliability was moderate (k = 0.56-0.59). Field classifications showed almost perfect agreement (k = 0.87-0.88), supporting operational applicability. ConclusionsThe Polovodova method provides robust estimates of parity and early gonotrophic-cycle status in Anopheles mosquitoes, but late-cycle classifications are more prone to reader disagreement and systematic underestimation. Operational use should therefore rely on trained multiple-reader workflows, including adjudication of discordant or late-cycle specimens, rather than single-reader classification. These findings provide empirical benchmarks for quality-assured Polovodova age-grading in malaria vector surveillance and intervention evaluation.

ecology↗

bioRxiv2026

TAUS: Target-Age Unified Survival. Survival analysis without assuming proportional hazards or parameterising the survival function.

O_LIStandard survival analysis methods often rely on the assumption of proportional hazards (PH) or parameterisations of the survival function that might not be appropriate for wild populations. C_LIO_LITo enable survival analysis without these modelling constraints, we developed an approach that combines the Kaplan-Meier estimator with conditional probability theory to compute age-specific probabilities of survival up to some target age of choice{tau} . Marginalising this probability over the age distribution of the population yields O{tau}, the probability that a randomly sampled individual of unknown age will outlive the target age{tau} . Notably, the value for{tau} is set by the analyst for each group independently, which allows accounting for differences in pace of life across populations. C_LIO_LIWe tested its application using a simulation study and two real-world datasets, and compared its performance against that of Cox PH and parametric survival models. The PH assumption was violated in the three examples, rendering the Cox PH models inappropriate. Parametric models offered a better alternative, but the best parametric fit missed at least some key survival patterns in all examples. The TAUS model provided a valid description of survival patterns in all cases. Its richer output also allowed finer analysis of survival differences between populations. C_LIO_LIThe TAUS model is also available as an R package (https://github.com/casasgomezuribarri/TAUS). This new approach to survival analysis without PH or parametric assumptions allows the comparison of survival probabilities across populations with different age structures and rates of pace of life. This makes it suitable for a wide range of ecological applications, including in population viability analysis, epidemiology, or life-history theory C_LI

ecology↗

bioRxiv2026

Ecology of mosquitoes in Scottish wetlands and confirmation of Culex molestus and hybrids in Scotland

BackgroundThe recent spread of mosquito-borne pathogens and their vectors within Europe highlights the impact of climate change on vector-borne disease (VBD) distributions. Mosquito surveillance has been implemented in many European countries to monitor expansion of vector populations and VBDs, but ability to predict disease risk is constrained by geographic data gaps, particularly in northern areas. In the United Kingdom, wetland mosquito surveillance has been conducted extensively in England, with a knowledge deficit for Scotland. Here, we addressed this gap through a nationwide survey of mosquitoes at Scottish wetlands, with aims of 1) confirming the geographic distribution and environmental drivers of mosquito occurrence and abundance, and 2) identifying the presence of vector species of epidemiological concern, including the Culex pipiens molestus biotype, an important vector of emerging VBDs in mainland Europe. MethodsMonthly mosquito sampling was conducted between June and October 2023 at 22 sites across Scotland comprising six wetland types: coastal saltmarsh, wet grassland, wet woodland, reedbeds, ponds and blanket bog. Adult and larval populations were sampled using Biogents BG-Pro traps and larval dipping respectively. Microclimatic and hydrological variables were recorded at collection sites and used in generalised linear mixed models to identify predictors of mosquito presence and abundance. Results1951 adults (17 species/ groups) and 860 larvae (six species/ groups) were collected from wetlands over 183 and 164 sampling events respectively. Mosquitoes were widely distributed across the Scottish mainland including up to the northern coast, being found at all but one site. Several potential vector species including Culex pipiens s.l. and Anopheles claviger were common. Amongst the adult Culex pipiens s.l. specimens, approximately 8% were Culex pipiens biotype molestus or hybrid forms. Total mosquito abundance and that of key vector species were positively associated with temperature and rainfall. ConclusionsWe report the widespread distribution of mosquitoes in wetlands throughout Scotland, including potential vector species previously unconfirmed in Scotland. Predicted associations between mosquito abundance, rainfall and temperature indicate that climate change could favour mosquito populations in Scotland. Our results provide the first comprehensive description of mosquito ecology in Scotland, as required to update assessment of VBD risk under climate change.

ecology↗

bioRxiv2025

Characterization of a new laboratory colony of Anopheles funestus mosquitoes established in Ifakara, Tanzania

BackgroundAnopheles funestus, a major vector of malaria in Africa, has proven difficult to colonize in laboratory settings, impeding research on its biology and control. After several attempts, our team recently succeeded in colonizing a strain of An. funestus from Tanzania (FUTAZ). The objective of this study was to analyse the key fitness and genotypic characteristics of these mosquitoes during multiple filial generations of laboratory adaptation and compare them to wild An. funestus from Tanzania and a pre-existing colony of An. funestus from Mozambique (FUMOZ). MethodsMeasures of mating success (percentage of female mosquitoes inseminated), body size (wing length), fecundity (number of eggs laid per female), and insecticide susceptibility (percentage of 24-hour mortality after exposure to insecticides) were compared between the newly established colonies of Tanzanian An. funestus (FUTAZ colonies), the long-established FUMOZ colonies, and a colony of Anopheles arabiensis maintained in the same laboratory. The maternal lineages of the An. funestus mosquitoes were investigated through a hydrolysis probe analysis of their mitochondrial DNA to identify distinct clades, I and II. Additionally, other intragenomic variations were examined through a PCR analysis of restriction fragment length polymorphisms (RFLP) on the third domain of 28S ribosomal DNA. These molecular markers were used to compare the FUTAZ colonies, FUMOZ colonies in Tanzania and South Africa, and the wild-collected An. funestus from Tanzania. ResultThe mating success and body size of FUTAZ females declined significantly from filial generations F1 to F6 relative to the founder population (F0), but then increased from F7 onwards eventually matching FUMOZ by F9. Fecundity was similar across all colonies tested. However, it took significantly longer for 50% of the females in the FUTAZ and FUMOZ colonies (over 10 days) to mate compared to females in the An. arabiensis colony (approximately 5 days). Insecticide resistance appeared to be lost during colonization, but this varied with insecticide classes. Majority of mosquitoes in the FUTAZ colony, as well as the wild-caught Tanzanian An. funestus belonged to Clade I (80.4-89.4%) and RFLP type "Y" (90.5-91.4%), while the FUMOZ colonies were mostly Clade II (65.5-88.5%) and RFLP type "MW" (90.5-91.5%). ConclusionThis study suggests that the mating success and body size of An. funestus decreases significantly during the early stages of colonization, then increase as the mosquitoes adapt to laboratory conditions. It is therefore crucial to have a large enough founder population to persist through these early generations in order to achieve stable colonization of An. funestus. The Clade and RFLP genotyping demonstrated the genetic similarities between the FUTAZ mosquitoes and wild-caught Tanzanian An. funestus, but also showed that the new colony can be distinguished from the FUMOZ colony.

ecology↗

bioRxiv2025

Phenology underlies apparent urbanisation effects on avian malaria in juvenile songbirds

Urbanisation can modify species interactions, including those between parasites and their hosts. In birds, urbanisation can either increase or decrease avian malaria infection, depending on host species, parasite or study location. However, temporal coordination between parasites and hosts, which may impact infection outcomes, has not been studied in urban ecology. To fill this gap, we collected blood samples from wild blue tit nestlings (Cyanistes caeruleus) in urban and forest habitats to examine how their hatch dates affected the prevalence and intensity of malaria infection. To separate parasites and quantify parasite load, we newly developed a species-specific qPCR assay. We found that Leucocytozoon prevalence was strongly affected by urbanisation, but effects depended on study year. This was driven by hatch date: nestlings that hatched earlier in the spring had a lower probability of being infected, independent of habitat type. In the few heavily infected nestlings, intensity of infection was associated with low body weight, suggesting fitness effects of infection. These results highlight the importance of breeding early to avoid early-life infection with malaria parasites, and that apparent urbanisation effects on infection arose from phenological differences between urban and forest habitats. Underappreciated phenological changes may underline other ecological effects of urbanisation.

ecology↗

bioRxiv2025

Towards Scalable Age-Grading of Aedes albopictus mosquito using Mid-Infrared Spectroscopy and Machine Learning

The age structure and dynamics of mosquito populations are crucial for understanding their ability to spread diseases and assessing the effectiveness of anti-mosquito control measures. However, available methods to age-grade mosquito populations are labour-intensive and imprecise, particularly for Aedes species. We investigated the potential of Mid-Infrared Spectroscopy (MIRS) combined with Supervised Machine Learning (ML) to rapidly and accurately predict the age of adult females and males of the arbovirus vector, Aedes albopictus. First, we demonstrated the ability of MIRS-ML to age male and female mosquitoes reared under laboratory conditions. Second, we optimised the model with adults emerged from wild collected eggs reared under natural conditions in a semi-field facility, to expose them to more realistic ambient conditions. For each sex we developed three ML models based on the resolution of the predicted adult age class: low (9 day interval), medium (6 days) and high resolution (3 days) from 1 to 15 or 33 days for males and females, respectively. The prediction accuracy decreased as the resolution increased. In males, the accuracy dropped from 99% (low) to 93% (medium) and 85.8% (low); in females the high and medium resolution models showed 89.4% and 78.5% accuracy, which decreased to 72.6% for the low resolution. In a simulated vector control intervention, the low-resolution models allowed to detect shifts in the age-structure of Ae. albopictus populations with minimal sampling effort (<100 specimens). Finally, we validated MIRS-ML on two unseen data and reconstructed plausible age structures in 1) laboratory-reared and 2) field-collected Ae. albopictus males and females. Overall, the results represent a first step towards the development of a sound and reproducible MIRS-ML approach for age-grading of Ae. albopictus populations in the wild. AUTHOR SUMMARYKnowing the age of mosquito populations is critical for understanding how effectively they can transmit viruses like dengue, chikungunya, and Zika, as older mosquitoes are more likely to be infectious. Also, comparing ages of mosquito population before and after a control intervention - such as insecticide aerial spraying - may allow to understand the impact of the intervention. However, existing methods to estimate mosquito age are time-consuming and imprecise. In this study, we tested whether a rapid and scalable method based on detection of age-related changes by mid-infrared spectroscopy (MIRS) coupled with machine learning (ML) could accurately estimate the age of Aedes albopictus, the Asian Tiger mosquito, an important arbovirus vector. We trained MIRS-ML models using mosquitoes reared in both laboratory and semi-field conditions to reflect realistic environmental variation. Our models were able to classify mosquito age with high accuracy, especially when grouping individuals into broader age categories. In simulated vector control scenarios, low-resolution models effectively detected shifts in population age structure with minimal sampling effort. We also applied our approach to field-collected mosquitoes that showed plausible age structures, suggesting potential of this approach for real-world surveillance. This method represents a promising, scalable, and non-destructive tool for monitoring mosquito population dynamics and could help monitor control strategies against Aedes-borne diseases.

ecology↗

bioRxiv2025

Distinct Genetic Populations and Resistance Backgrounds of the Malaria Vector Anopheles funestus in Tanzania

Population genetic analysis of mosquitoes is becoming increasingly important for understanding the distribution of insecticide resistance alleles, devising sustainable insecticide-based vector control approaches, and how malaria vector populations are structured in space. Anopheles funestus is the dominant malaria vector in Tanzania and most parts of East and Southern Africa. To better understand its population genomic structure in Tanzania, we sequenced the genomes of 334 individual An. funestus mosquitoes from 11 administrative regions. We found two genetically differentiated populations; one inland and at high altitude (found in Katavi, Kagera, Kigoma, and Mwanza) and a second coastal, at low altitude (found in Pwani, Morogoro, Tanga, Ruvuma, Mtwara, Dodoma, and Lindi), with differences in genetic diversity and inbreeding. We found asynchronous selective sweeps, associated with insecticide resistance phenotypes, at the Cyp9k1 gene, and Cyp6p gene cluster, with distinct copy number-variant profiles between the coastal and inland populations. These results suggest that inland and coastal An. funestus populations have divergent histories, with the arid, central region of Tanzania, which also contains the Rift Valley being a possible barrier to gene flow. Such population disconnectedness should be considered for insecticide deployment, resistance management, and the rollout of novel genetic- based vector control approaches. These findings provide the most detailed study of Tanzanian An. funestus population structure and resistance genetics to date. Future research should examine the epidemiological relevance of this discontinuity in gene flow and whether these populations have different malaria transmission abilities.

genetics↗

bioRxiv2025

Wavefront estimation through structured detection in laser scanning microscopy

Laser scanning microscopy (LSM) is the base of numerous advanced imaging techniques, including confocal laser scanning microscopy (CLSM), a widely used tool in life sciences research. However, its effective resolution is often compromised by optical aberrations, a common challenge in all optical systems. While adaptive optics (AO) can correct these aberrations, current methods face significant limitations: Aberration estimation, which is central to any AO approach, typically requires specialized hardware or prolonged sample exposure, rendering these methods sample-invasive, and less user-friendly. In this study, we introduce a simple and efficient AO approach for CLSM systems equipped with a detector array - the same of super-resolved image-scanning microscopy - and an AO element for beam shaping. We demonstrate that imaging datasets acquired with a detector array inherently encode aberration information. Leveraging this property, we developed a custom convolutional neural network capable of decoding aberrations, up to the 11th Zernike coefficient, directly from a single acquisition. This method enables a new generation of AO implementations for LSM, offering an accessible solution that minimizes sample stress while achieving high-resolution, and aberration-free imaging.

bioengineering↗

bioRxiv2024

Discovery of knock-down resistance in the major malaria vector Anopheles funestus reveals the legacy of persistent DDT pollution.

A major mechanism of insecticide resistance in insect pests is knock-down resistance (kdr) caused by mutations in the voltage-gated sodium channel (Vgsc) gene. Despite being common in most malaria Anopheles vector species, kdr mutations have never been observed in Anopheles funestus, the principal malaria vector in Eastern and Southern Africa. While monitoring 10 populations of An. funestus in Tanzania, we unexpectedly found resistance to DDT, a banned insecticide, in one location. Through whole-genome sequencing of 333 An. funestus samples from these populations, we found 8 novel amino acid substitutions in the Vgsc gene, including the kdr variant, L976F (L1014F in An. gambiae), in tight linkage disequilibrium with another (P1842S). The mutants were found only at high frequency in one region, with a significant decline between 2017 and 2023. Notably, kdr L976F was strongly associated with survivorship to the exposure to DDT insecticide, while no clear association was noted with a pyrethroid insecticide (deltamethrin). Further study is necessary to identify the origin and spread of kdr in An. funestus, and the potential threat to current insecticide-based vector control in Africa. SignificanceKnock-down resistance (kdr) mutations confer resistance to malaria vector control insecticides and pose a grave threat to malaria control. Here, we report the first discovery of kdr in An. funestus, the principal malaria vector in East and Southern Africa. Kdr in An. funestus conferred resistance to DDT but not deltamethrin. Based on extensive DDT contamination and unofficial usage in Tanzania, it is possible that kdr emerged because of widespread organic pollution as opposed to through public health efforts. Regardless of origin, the discovery of kdr in An. funestus is an alarming development that warrants immediate, extensive follow-up and close surveillance to establish the origin, and extent to which it may threaten malaria control in An. funestus.

genetics↗

bioRxiv2024

Advancing age grading techniques for Glossina morsitans morsitans, vectors of African trypanosomiasis, through mid-infrared spectroscopy and machine learning

Tsetse are the insects responsible for transmitting African trypanosomes, which cause sleeping sickness in humans and animal trypanosomiasis in wildlife and livestock. Knowing the age of these flies is important when assessing the effectiveness of vector control programs and modelling disease risk. However, current methods to assess fly age are labour-intensive, slow, and often inaccurate as skilled personnel are in short supply. Mid-infrared spectroscopy (MIRS), a fast and cost-effective tool to accurately estimate several biological traits of insects, offers a promising alternative. This is achieved by characterising the biochemical composition of the insect cuticle using infrared light coupled with machine learning algorithms to estimate the traits of interest. We tested the performance of MIRS in estimating tsetse sex and age for the first time using spectra obtained from their cuticle. We used 541 insectary-reared Glossina m. morsitans of two different age groups for males (5 and 7 weeks) and three age groups for females (3 days, 5 weeks, and 7 weeks). Spectra were collected from the head, thorax, and abdomen of each sample. Machine learning models differentiated between male and female flies with a 96% accuracy and predicted the age group with 94% and 87% accuracy for males and females, respectively. The key infrared regions important for discriminating sex and age classification were characteristic of lipid and protein content. Our results support the use of MIRS as a fast and accurate way to identify tsetse sex and age with minimal pre-processing. Further validation using wild-caught tsetse can pave the way for this technique to be implemented as a routine surveillance tool in vector control programmes. Author summaryMale and female tsetse transmit the parasites that cause sleeping sickness in humans and nagana in livestock. To control these diseases, knowing the age of these flies is important, as it helps evaluate the efficacy of control measures and assess disease risk. However, current age-grading methods are laborious, often unreliable, and in the case of male tsetse, highly inaccurate. This study explores a novel approach that uses mid-infrared spectroscopy (MIRS) to estimate the age of individual tsetse. Machine learning can detect signatures in MIRS that help identify the composition of a flys cuticle, which differs between sexes and changes as they age. We trained machine learning models that distinguished male from female flies with 96% accuracy and predicted the correct age group with 94% accuracy for males and 87% accuracy for females. MIRS offers a fast and reliable way to identify tsetse sex and age with minimal preparation. If this method is successfully validated with wild flies, it holds the potential to vastly increase the accuracy of the way we monitor and combat these disease-carrying insects, thus offering significant advantages in our efforts to control them.

biochemistry↗

bioRxiv2023

The Salivary Microbiome and Predicted Metabolite Production are Associated with Progression from Barrett's Esophagus to Esophageal Adenocarcinoma

Esophageal adenocarcinoma (EAC) is rising in incidence and associated with poor survival, and established risk factors do not explain this trend. Microbiome alterations have been associated with progression from the precursor Barretts esophagus (BE) to EAC, yet the oral microbiome, tightly linked to the esophageal microbiome and easier to sample, has not been extensively studied in this context. We aimed to assess the relationship between the salivary microbiome and neoplastic progression in BE to identify microbiome-related factors that may drive EAC development. We collected clinical data and oral health and hygiene history and characterized the salivary microbiome from 250 patients with and without BE, including 78 with advanced neoplasia (high grade dysplasia or early adenocarcinoma). We assessed differential relative abundance of taxa by 16S rRNA gene sequencing and associations between microbiome composition and clinical features and used microbiome metabolic modeling to predict metabolite production. We found significant shifts and increased dysbiosis associated with progression to advanced neoplasia, with these associations occurring independent of tooth loss, and the largest shifts were with the genus Streptococcus. Microbiome metabolic models predicted significant shifts in the metabolic capacities of the salivary microbiome in patients with advanced neoplasia, including increases in L- lactic acid and decreases in butyric acid and L-tryptophan production. Our results suggest both a mechanistic and predictive role for the oral microbiome in esophageal adenocarcinoma. Further work is warranted to identify the biological significance of these alterations, to validate metabolic shifts, and to determine whether they represent viable therapeutic targets for prevention of progression in BE.

microbiology↗

bioRxiv2022

Using transfer learning and dimensionality reduction techniques to improve generalisability of machine-learning predictions of mosquito ages from mid-infrared spectra

Accurate prediction of mosquito population age structures can improve the evaluation of mosquito-targeted interventions since old mosquitoes are more likely to transmit malaria than young ones. Mid-infrared spectroscopy (MIRS) reveals age-associated variation in the biochemical composition of the mosquito cuticle, which can then be used to train machine learning (ML) models to predict mosquito ages. However, these MIRS-ML models are not always generalisable across different mosquito populations. Here, we investigated whether dimensionality reduction applied to the MIRS input data and transfer learning could improve the generalisability of MIRS-ML predictions for mosquito ages. We reared adults of the malaria vector, Anopheles arabiensis, in two insectaries (Ifakara, Tanzania and Glasgow, UK). The heads and thoraces of female mosquitoes of two age classes (1-9 day-olds and 10-17 day-olds) were scanned using an attenuated total reflection-Fourier transform infrared (ATR-FTIR) spectrometer (4000 cm-1 to 400 cm-1). The dimensionality of the spectra data was reduced using unsupervised principal component analysis (PCA) or t-distributed stochastic neighbour embedding (t-SNE), and then the spectra were used to train deep learning (DL) and standard machine learning (ML) classifiers. Transfer learning was also evaluated for improving the computational cost of the models when predicting mosquito age classes from new populations. Model accuracies for predicting the age of test mosquitoes from the same insectary as the training samples reached 99% for DL and 92% for ML, but did not generalise to a different insectary, achieving only 46% and 48% for ML for DL, respectively. Dimensionality reduction did not improve the model generalisability between locations but reduced computational time up to 5-fold. However, transfer learning by updating pre-trained models with 2% of mosquitoes from the alternate location brought both DL and standard ML model performance to ~98% accuracy for predicting mosquito age classes in the alternative insectary. Combining dimensionality reduction and transfer learning can reduce computational costs and improve the transferability of both deep learning and standard machine learning models for predicting the age of mosquitoes. Future studies could investigate the optimal quantities and diversity of training data necessary for transfer learning, and implications for broader generalisability to unseen datasets.

bioinformatics↗

bioRxiv2021

Preterm birth is associated with xenobiotics and predicted by the vaginal metabolome

Spontaneous preterm birth (sPTB) is a leading cause of maternal and neonatal morbidity and mortality, yet both its prevention and early risk stratification are limited. The vaginal microbiome has been associated with PTB risk, possibly via metabolic or other interactions with its host. Here, we performed untargeted metabolomics on 232 vaginal samples, in which we have previously profiled the microbiota using 16S rRNA gene sequencing. Samples were collected at 20-24 weeks of gestation from women with singleton pregnancies, of which 80 delivered spontaneously before 37 weeks of gestation. We find that the vaginal metabolome correlates with the microbiome and separates into six clusters, three of which are associated with spontaneous preterm birth (sPTB) in Black women. Furthermore, while we identify five metabolites that associate with sPTB, another five associate with sPTB only when stratifying by race. We identify multiple microbial correlations with metabolites associated with sPTB, including intriguing correlations between vaginal bacteria that are considered sub-optimal and metabolites that were enriched in women who delivered at term. We propose that several sPTB-associated metabolites may be exogenous, and investigate another using metabolic models. Notably, we use machine learning models to predict sPTB risk using metabolite levels, weeks to months in advance, with high accuracy. We show that these predictions are more accurate than microbiome-based and maternal covariates-based models. Altogether, our results demonstrate the potential of vaginal metabolites as early biomarkers of sPTB and highlight exogenous exposures as potential risk factors for prematurity.

systems biology↗

bioRxiv2020

Rapid ageing and species identification of natural mosquitoes for malaria surveillance

The malaria parasite, which is transmitted by several Anopheles mosquito species, requires more time to reach its human-transmissible stage than the average lifespan of a mosquito. Monitoring the species-specific age structure of mosquito populations is critical to evaluating the impact of vector control interventions on malaria risk. We developed a rapid, cost-effective surveillance method based on deep learning of mid-infrared spectra of mosquitoes cuticle that simultaneously identifies the species and the age of three main malaria vectors, in natural populations. Using over 40,000 ecologically and genetically diverse females, we could speciate and age grade An. gambiae, An. arabiensis, and An. coluzzii with up to 95% accuracy. Further, our model learned the age of new populations with minimal sampling effort and detected the impact of control interventions on simulated mosquito populations, measured as a shift in their age structures. We anticipate our method to be applied to other arthropod vector-borne diseases.

zoology↗

bioRxiv2019

Parkinson’s disease-associated alterations of the gut microbiome can invoke disease-relevant metabolic changes

Parkinsons disease (PD) is a systemic disease clinically defined by the degeneration of dopaminergic neurons in the brain. While alterations in the gut microbiome composition have been reported in PD, their functional consequences remain unclear. Herein, we first analysed the gut microbiome of patients and healthy controls by 16S rRNA gene sequencing of stool samples from the Luxembourg Parkinsons study (n=147 typical PD cases, n=162 controls). All individuals underwent detailed clinical assessment, including neurological examinations and neuropsychological tests followed by self-reporting questionnaires. Second, we predicted the potential secretion for 129 microbial metabolites through personalised metabolic modelling using the microbiome data and genome-scale metabolic reconstructions of human gut microbes. Our key results include: 1. eight genera and nine species changed significantly in their relative abundances between PD patients and healthy controls. 2. PD-associated microbial patterns statistically depended on sex, age, BMI, and constipation. The relative abundances of Bilophila and Paraprevotella were significantly associated with the Hoehn and Yahr staging after controlling for the disease duration. In contrast, dopaminergic medication had no detectable effect on the PD microbiome composition. 3. Personalised metabolic modelling of the gut microbiomes revealed PD-associated metabolic patterns in secretion potential of nine microbial metabolites in PD, including increased methionine and cysteinylglycine. The microbial pantothenic acid production potential was linked to the presence of specific non-motor symptoms and attributed to individual bacteria, such as Akkermansia muciniphila and Bilophila wardswarthia. Our results suggest that PD-associated alterations of gut microbiome could translate into functional differences affecting host metabolism and disease phenotype.

microbiology↗

bioRxiv2019

Detection of malaria in insectary-reared Anopheles gambiae using near-infrared spectroscopy

Large-scale surveillance of mosquito populations is crucial to assess the intensity of vector-borne disease transmission and the impact of control interventions. However, there is a lack of accurate, cost-effective and high-throughput tools for mass-screening of vectors. This study demonstrates proof-of-concept that near-infrared spectroscopy (NIRS) is capable of rapidly identifying laboratory strains of human malaria infection in African mosquito vectors. By using partial least square regression models based on malaria-infected and uninfected Anopheles gambiae mosquitoes, we showed that NIRS can detect oocyst- and sporozoite-stage Plasmodium falciparum infections with 88% and 95% accuracy, respectively. Accurate, low-cost, reagent-free screening of mosquito populations enabled by NIRS could revolutionize surveillance and elimination strategies for the most important human malaria parasite in its primary African vector species. Further research is needed to evaluate how the method performs in the field following adjustments in the training datasets to include data from wild-caught infected and uninfected mosquitoes.

epidemiology↗

bioRxiv2018

Prediction of malaria mosquito species and population age structure using mid-infrared spectroscopy and supervised machine learning

Despite the global efforts made in the fight against malaria, the disease is resurging. One of the main causes is the resistance that Anopheles mosquitoes, vectors of the disease, have developed to insecticides. Anopheles must survive for at least 12 days to possibly transmit malaria. Therefore, to evaluate and improve malaria vector control interventions, it is imperative to monitor and accurately estimate the age distribution of mosquito populations as well as total population sizes. However, estimating mosquito age is currently a slow, imprecise, and labour-intensive process that can only distinguish under-from over-four-day-old female mosquitoes. Here, we demonstrate a machine-learning based approach that utilizes mid-infrared spectra of mosquitoes to characterize simultaneously, and with unprecedented accuracy, both age and species identity of females of the malaria vectors Anopheles gambiae and An. arabiensis mosquitoes within their respective populations. The prediction of the age structures was statistically indistinguishable from true modelled distributions. The method has a negligible cost per mosquito, does not require highly trained personnel, is substantially faster than current techniques, and so can be easily applied in both laboratory and field settings. Our results show that, with larger mid-infrared spectroscopy data sets, this technique can be further improved and expanded to vectors of other diseases such as Zika and Dengue.

bioinformatics↗