Stratifying risk of disease in haematuria patients using machine learning techniques to improve diagnostics.

Background: Detailed and invasive clinical investigations are required to identify the causes of haematuria. Highly unbalanced patient population (predominantly male) and a wide range of potential causes make the ability to correctly classify patients and identify patient-specific biomarkers a major challenge. Studies have shown that it is possible to improve the diagnosis using multi-marker analysis, even in unbalanced datasets, by applying advanced analytical methods. Here, we applied several machine learning algorithms to classify patients from the haematuria patient cohort (HaBio) by analysing multiple biomarkers and to identify the most relevant ones. Materials and methods: We applied several classification and feature selection methods (k-means clustering, decision trees, random forest with LIME explainer and CACTUS algorithm) to stratify patients into two groups: healthy (with no clear cause of haematuria) or sick (with an identified cause of haematuria e.g., bladder cancer, or infection). The classification performance of the models was compared. Biomarkers identified as important by the algorithms were also analysed in relation to their involvement in the pathological processes. Results: Results showed that a high unbalance in the datasets significantly affected the classification by random forest and decision trees, leading to the overestimation of the sick class and low model performance. CACTUS algorithm was more robust to the unbalance in the dataset. CACTUS obtained a balanced accuracy of 0.747 for both genders, 0.718 for females and 0.803 for males. The analysis showed that in the classification process for the whole dataset: microalbumin, male gender, and tPSA emerged as the most informative biomarkers. For males: age, microalbumin, tPSA, cystatin C, BTA, HAD and S100A4 were the most significant biomarkers while for females microalbumin, IL-8, pERK, and CXCL16. Conclusions: CACTUS algorithm demonstrated improved performance compared with other methods such as decision trees and random forest. Additionally, we identified the most relevant biomarkers for the specific patient group, which could be considered in the future as novel biomarkers for diagnosis. Our results have the potential to inform future research and provide new personalised diagnostic approaches tailored directly to the needs of the individuals.

Exploring the antimicrobial peptidome of nematodes through phylum-spanning in silico analyses highlights novel opportunities for pathogen control.

Antimicrobial Peptides (AMPs) are key constituents of the invertebrate innate immune system and provide critical protection against microbial threat. Nematodes display diverse life strategies where they are exposed to heterogenous, microbe rich, environments highlighting their need for an innate immune system. Within the Ecdysozoa, arthropod AMPs have been well characterised, however nematode-derived AMP knowledge is limited. In this study the distribution and abundance of putative AMP-encoding genes was examined in 134 nematode genomes providing the most comprehensive profile of AMP candidates within phylum Nematoda. Through genome and transcriptome analyses we reveal that phylum Nematoda is a rich source of putative AMP diversity and demonstrate (i) putative AMP group profiles that are influenced by nematode lifestyle where free-living nematodes appear to display enriched putative AMP profiles relative to parasitic species; (ii) major differences in the putative AMP profiles between nematode clades where Clade 9/V and 10/IV species possess expanded putative AMP repertoires; (iii) AMP groups with highly restricted profiles (e.g. Cecropins and Diapausins) and others [e.g. Nemapores and Glycine Rich Secreted Peptides (GRSPs)] which are more widely distributed; (iv) complexity in the distribution and abundance of CSαß subgroup members; and (v) that putative AMPs are expressed in host-facing life stages and biofluids of key nematode parasites. These data indicate that phylum Nematoda displays diversity in putative AMPs and underscores the need for functional characterisation to reveal their role and importance to nematode biology and host-nematode-microbiome interactions.

Novel integrated computational AMP discovery approaches highlight diversity in the helminth AMP repertoire.

Antimicrobial Peptides (AMPs) are immune effectors that are key components of the invertebrate innate immune system providing protection against pathogenic microbes. Parasitic helminths (phylum Nematoda and phylum Platyhelminthes) share complex interactions with their hosts and closely associated microbiota that are likely regulated by a diverse portfolio of antimicrobial immune effectors including AMPs. Knowledge of helminth AMPs has largely been derived from nematodes, whereas the flatworm AMP repertoire has not been described. This study highlights limitations in the homology-based approaches, used to identify putative nematode AMPs, for the characterisation of flatworm AMPs, and reveals that innovative algorithmic AMP prediction approaches provide an alternative strategy for novel helminth AMP discovery. The data presented here: (i) reveal that flatworms do not encode traditional lophotrochozoan AMP groups (Big Defensin, CSαß peptides and Myticalin); (ii) describe a unique integrated computational pipeline for the discovery of novel helminth AMPs; (iii) reveal >16,000 putative AMP-like peptides across 127 helminth species; (iv) highlight that cysteine-rich peptides dominate helminth AMP-like peptide profiles; (v) uncover eight novel helminth AMP-like peptides with diverse antibacterial activities, and (vi) demonstrate the detection of AMP-like peptides from Ascaris suum biofluid. These data represent a significant advance in our understanding of the putative helminth AMP repertoire and underscore a potential untapped source of antimicrobial diversity which may provide opportunities for the discovery of novel antimicrobials. Further, unravelling the role of endogenous worm-derived antimicrobials and their potential to influence host-worm-microbiome interactions may be exploited for the development of unique helminth control approaches.

Phylum-Spanning Neuropeptide GPCR Identification and Prioritization: Shaping Drug Target Discovery Pipelines for Nematode Parasite Control.

Nematode parasites undermine human health and global food security. The frontline anthelmintic portfolio used to treat parasitic nematodes is threatened by the escalation of anthelmintic resistance, resulting in a demand for new drug targets for parasite control. Nematode neuropeptide signalling pathways represent an attractive source of novel drug targets which currently remain unexploited. The complexity of the nematode neuropeptidergic system challenges the discovery of new targets for parasite control, however recent advances in parasite 'omics' offers an opportunity for the in silico identification and prioritization of targets to seed anthelmintic discovery pipelines. In this study we employed Hidden Markov Model-based searches to identify ~1059 Caenorhabditis elegans neuropeptide G-protein coupled receptor (Ce-NP-GPCR) encoding gene homologs in the predicted protein datasets of 10 key parasitic nematodes that span several phylogenetic clades and lifestyles. We show that, whilst parasitic nematodes possess a reduced complement of Ce-NP-GPCRs, several receptors are broadly conserved across nematode species. To prioritize the most appealing parasitic nematode NP-GPCR anthelmintic targets, we developed a novel in silico nematode parasite drug target prioritization pipeline that incorporates pan-phylum NP-GPCR conservation, C. elegans-derived reverse genetics phenotype, and parasite life-stage specific expression datasets. Several NP-GPCRs emerge as the most attractive anthelmintic targets for broad spectrum nematode parasite control. Our analyses have also identified the most appropriate targets for species- and life stage- directed chemotherapies; in this context we have identified several NP-GPCRs with macrofilaricidal potential. These data focus functional validation efforts towards the most appealing NP-GPCR targets and, in addition, the prioritization strategy employed here provides a blueprint for parasitic nematode target selection beyond NP-GPCRs.

Ascaris suum Informs Extrasynaptic Volume Transmission in Nematodes.

Neural circuit synaptic connectivities (the connectome) provide the anatomical foundation for our understanding of nematode nervous system function. However, other nonsynaptic routes of communication are known in invertebrates including extrasynaptic volume transmission (EVT), which enables short- and/or long-range communication in the absence of synaptic connections. Although EVT has been highlighted as a facet of Caenorhabditis elegans neurosignaling, no experimental evidence identifies body cavity fluid (pseudocoelomic fluid; PCF) as a vehicle for either neuropeptide or biogenic amine transmission. In the parasitic nematode Ascaris suum, FMRFamide-like peptides encoded on flp-18 potently stimulate female reproductive organs but are expressed in cells that are anatomically distant from the reproductive organ, with no known synaptic connections to this tissue. Here we investigate nonsynaptic neuropeptide signaling in nematodes mediated by the body cavity fluid. Our data show that (i) A. suum PCF (As-PCF) contains a catalog of neuropeptides including FMRFamide-like peptides and neuropeptide-like proteins, (ii) the A. suum FMRFamide-like peptide As-FLP-18A dominates the As-PCF peptidome, (iii) As-PCF potently modulates nematode reproductive muscle function ex vivo, mirroring the effects of synthetic FLP-18 peptides, (iv) As-PCF activates the C. elegans FLP-18 receptors NPR-4 and -5, (v) As-PCF alters C. elegans behavior, and (vi) FLP-18 and FLP-18 receptors display pan-phylum distribution in nematodes. This study provides the first direct experimental evidence to support an extrasynaptic volume route for neuropeptide transmission in nematodes. These data indicate nonsynaptic signaling within the nematode functional connectome and are particularly pertinent to receptor deorphanization approaches underpinning drug discovery programs for nematode pathogens.