Biology subjects

Saccenti, E.

Publications and source records attributed to Saccenti, E..

14 recordsLinked to original sources

bioRxiv2025

MiGenPro: A linked data workflow for phenotype-genotype prediction of microbial traits using machine learning.

The availability of microbial genomic data and the development of machine learning methods have created a unique opportunity to establish associations between genetic information and phenotypes. Here, we introduce a computational workflow for Microbial Genome Prospecting (MiGenPro) that combines phenotypic and genomic information. MiGenPro serves as a workflow for the training of machine learning models that predict microbial traits from genomes that have been annotated. Microbial genomes have been consistently annotated and features were stored in a semantic framework that is easy to query using SPARQL. The data was used to train machine learning models and successfully predicted microbial traits such as motility, Gram stain, oxygen requirement, optimal temperature range, and sporulation capabilities. To ensure robustness, a hyperparameter halving grid search was used to determine optimal parameter settings followed by five randomised train-test split validations which demonstrated consistent model performance across iterations and without overfitting. Effectiveness was further validated through comparison with existing models, showing comparable accuracy, with modest variations attributed to differences in datasets rather than methodology. Classification can be further explored using feature importance characterisation to identify biologically relevant genomic features. MiGenPro provides an easy to use interoperable workflow to build and validate models to predict phenotypes from microbes based on their annotated genome. Key MessagesO_LIMiGenPro merges phenotypic and genomic data for microbial trait prediction using machine learning. C_LIO_LIMiGenPro mitigates phenotype prediction model constraints by leveraging linked data technologies. C_LIO_LIMiGenPros FAIR design enables adaptation for various phenotypes with available training data. C_LI

bioinformatics↗

bioRxiv2025

Gene expression profile dynamics of earthworms exposed to ZnO and ZnO:Mn nanomaterials

Zinc oxide containing nanomaterials may elicit toxic responses in environmental organisms such as earthworms. Although toxic responses in earthworms have been reported, few studies have attempted to understand the molecular mode of action dynamics explaining the observed toxicity at a transcriptional level. This study investigates the time-dependent gene expression response in earthworms exposed to ZnO nanomaterials and ZnO:Mn multicomponent nanomaterials. Earthworms were exposed to ZnO or ZnO:Mn for 7 days and to ZnO:Mn or MnCl2 for 14 days. Strong differential gene expression responses were observed after 4 days of exposure to ZnO nanomaterials and after 2 days of exposure to ZnO:Mn. Moderate differential gene expression responses were observed after 14 days of exposure to ZnO:Mn and MnCl2. Gene ontology (GO) enrichment analysis revealed that differentially expressed genes in earthworms exposed to ZnO were associated with terms such as actin, (striated) muscle cells, contractile fiber, myofibril, sarcomere and supramolecular cellular components. In addition, genes that were upregulated after 2 days of exposure to ZnO:Mn were linked to GO terms including cilium, microtubule, cell projection, axoneme and sperm flagellum cellular components. Downregulated genes were enriched to GO terms related to ribosomes, mitochondria, translation, peptide processes, respiration and oxidative phosphorylation. Finally, for earthworms exposed to ZnO:Mn and MnCl2 for 14 days, only a limited number of differentially expressed genes were involved in GO terms related to diverse biological implications. In summary, exposing earthworms to the ZnO and ZnO:Mn nanomaterials elicited a transient response in differential gene expression related to muscle biology and energy metabolism and translation that had largely disappeared by day 14 from the start of the exposure.

systems biology↗

bioRxiv2025

Steering fatty acid composition of yeast microbial oil via genetic modification and bioprocess adjustment

The increasing demand for palm oil has drastic effects on the ecosystem as its production is not sustainable. To that end, developing a sustainable alternative to fatty acids and oils is urgent and of utmost interest. Oils produced by oleaginous yeasts present a promising solution, particularly because the fatty acid profile of the oil produced by these yeasts is comparable to that of plant-based oils and fats. The fatty acid composition of the oil determines its physiological properties, thereby determining its potential applications. Accordingly, the production of microbial oil with an optimal composition profile for a specific application is of great importance. In this study, we evaluated the variation that occurred in fatty acid composition due to different cultivation parameters (temperature, C/N ratio, carbon, and nitrogen sources) and applied genetic modifications to improve the lipid accumulation of Cutaneotrichosporon oleaginosus and Yarrowia lipolytica. We showed that specific fatty acid profiles associated with a particular application can be obtained by carefully selecting the microorganism and cultivation conditions.

microbiology↗

bioRxiv2024

Predicting uptake and elimination kinetics of chemicals in invertebrates: a technicalnote on residual variance modeling

Toxicokinetic models for predicting contents of nanomaterials and other toxic chemicals are often fitted without evaluation of the residual variance structure. The aim of the present study was to evaluate various residual variance structures, assuming either homoscedasticity or heteroscedasticity, when fitting non-linear toxicokinetic one-compartment models for predicting uptake, bioaccumulation and elimination of chemicals in invertebrate organisms. Data describing the exposure of several aquatic and terrestrial invertebrates to specific metal nanomaterials and other chemicals were available from real experiments for evaluating the residual variance functions for toxicokinetic models. As proof of concept, datasets of truly homoscedastic and heteroscedastic nature were simulated. Depending the dataset, applying models with different residuals variance assumption largely affected the residual plots and the error margins of parameters or the predicted content of a chemical. Consequently, selecting the most accurate residual variance functions for toxicokinetic modeling, either homoscedastic or heteroscedastic, improves the prediction of chemical contents in invertebrate organisms and the estimation of the associated uptake and elimination rates. HighlightsO_LIResidual plots indicate if an accurate model was fitted to the toxicological data C_LIO_LIChoice of residual variance function affects the error margins of predicted chemicals C_LIO_LISelecting proper residual variance models may prevent false positives and negatives C_LI

pharmacology and toxicology↗

bioRxiv2024

Novel multi-omics deconfounding variational autoencoders can obtain meaningful disease subtyping

Unsupervised learning, particularly clustering, plays a pivotal role in disease subtyp- ing and patient stratification, especially with the abundance of large-scale multi-omics data. Deep learning models, such as variational autoencoders (VAEs), can enhance clustering algorithms by leveraging inter-individual heterogeneity. However, the impact of confounders - external factors unrelated to the condition, e.g. batch effect or age - on clustering is often overlooked, introducing bias and spurious biological conclusions. In this work, we introduce four novel VAE-based deconfounding frameworks tailored for clustering multi-omics data. These frameworks effectively mitigate confounding effects while preserving genuine biological patterns. The deconfounding strategies employed include: i) removal of latent features correlated with confounders ii) a conditional variational autoencoder, iii) adversarial training, and iv) adding a regularization term to the loss function. Using real-life multi-omics data from TCGA, we simulated various confounding effects (linear, non-linear, categorical, mixed) and assessed model performance across 50 repetitions based on reconstruction error, clustering stability, and deconfounding efficacy. Our results demonstrate that our novel models, particularly the conditional multi-omics VAE (cXVAE), successfully handle simulated confounding effects and recover biologically-driven clustering structures. cXVAE accurately identifies patient labels and unveils meaningful pathological associations among cancer types, validating deconfounded representations. Furthermore, our study suggests that some of the proposed strategies, such as adversarial training, prove insufficient in confounder removal. In summary, our study contributes by proposing innovative frameworks for simultaneous multi-omics data integration, dimensionality reduction, and deconfounding in clustering. Benchmarking on open-access data offers guidance to end-users, facilitating meaningful patient stratification for optimized precision medicine.

genomics↗

bioRxiv2023

mEthAE: an Explainable AutoEncoder for methylation data

1In the quest to unravel the mysteries of our epigenetic landscape, researchers are continually challenged by the relationships among CpG sites. Traditional approaches are often limited by the immense complexity and high dimensionality of DNA methylation data. To address this problem, deep learning algorithms, such as autoencoders, are increasingly applied to capture the complex patterns and reduce dimensionality into latent space. In this pioneering study, we introduce an innovative chromosome-wise autoencoder, termed mEthAE, specifically designed for the interpretive reduction of methylation data. mEthAE achieves an impressive 400-fold reduction in data dimensions without compromising on reconstruction accuracy or predictive power in the latent space. In attempt to go beyond mere data compression, we developed a perturbation-based method for interpretation of latent dimensions. Through our approach we identified clusters of CpG sites that exhibit strong connections across all latent dimensions, which we refer to as global CpGs. Remarkably, these global CpGs are more frequently highlighted in epigenome-wide association studies (EWAS), suggesting our methods ability to pinpoint biologically significant CpG sites. Our findings reveal a surprising lack of correlation patterns, or even physical proximity on the chromosome among these connected CpGs. This leads us to propose an intriguing hypothesis: our autoencoder may be detecting complex, long-range, non-linear interaction patterns among CpGs. These patterns, largely uncharacterised in current epigenetic research, hold the potential to shed new light on our understanding of epigenetics. In conclusion, this study not only showcases the power of autoencoders in untangling the complexities of epigenetic data but also opens up new avenues for understanding the hidden connections within CpGs. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=168 SRC="FIGDIR/small/549496v2_ufig1.gif" ALT="Figure 1"> View larger version (31K): org.highwire.dtl.DTLVardef@16c5076org.highwire.dtl.DTLVardef@16b455org.highwire.dtl.DTLVardef@9948e8org.highwire.dtl.DTLVardef@181914b_HPS_FORMAT_FIGEXP M_FIG C_FIG

bioinformatics↗

bioRxiv2023

P-PPI: accurate prediction of peroxisomal protein-protein interactions (P-PPI) using deep learning-based protein sequence embeddings

Protein-protein interactions (PPIs) are crucial for various biological processes, and their prediction is typically accomplished through experimental methods, which can be time-consuming and costly. Computational methods provide a faster and more cost-effective approach, leveraging protein sequences and other data sources to infer PPIs. Deep learning (DL) approaches have shown promising results in various protein-related tasks, including PPI prediction. However, DL-based embeddings are often not thoroughly compared or evaluated against state-of-the-art tools. Additionally, existing PPI predictors incorporate different types of information beyond protein sequence representation, making it important to assess the effectiveness of DL-based embeddings solely relying on protein sequences. In this work, we benchmark and compare commonly used DL-based embeddings for PPI prediction based solely on protein sequence information. We utilize high-quality training data, including experimentally validated negative interactions from the Negatome database. The best model, obtained through double cross-validation and hyperparameter optimization, is selected and evaluated to predict peroxisomal PPIs. The resulting tool, P-PPI, is further enhanced by combining AlphaFold2-Multimer predictions with the P-PPI model, leveraging DL-based embeddings and protein structure predictions for a comprehensive analysis of peroxisomal PPIs. This integrated approach holds significant potential to advance our understanding of complex protein networks and their functions.

bioinformatics↗

bioRxiv2023

PortPred: exploiting deep learning embeddings of amino acid sequences for the identification of transporter proteins and their substrates

The physiology of every living cell is regulated at some level by transporter proteins which constitute a relevant portion of membrane-bound proteins and are involved in the movement of ions, small and macromolecules across bio-membranes. The importance of transporter proteins is unquestionable. The prediction and study of previously unknown transporters can lead to the discovery of new biological pathways, drugs and treatments. Here we present PortPred, a tool to accurately identify transporter proteins and their substrate starting from the protein amino acid sequence. PortPred successfully combines pre-trained deep learning-based protein embeddings and machine learning classification approaches and outperforms other state-of-the-art methods. In addition, we present a comparison of the most promising protein sequence embeddings (Unirep, SeqVec, ProteinBERT, ESM-1b) and their performances for this specific task.

bioinformatics↗

bioRxiv2022

OrganelX Web Server for Sub-Peroxisomal and Sub-Mitochondrial protein localisation

Computational approaches for sub-organelle protein localisation and identification are often neglected while general methods, not suitable for specific use cases, are promoted instead. In particular, organelle-specific research lacks user-friendly and easily accessible computational tools that allow researchers to perform computational analysis before starting time-consuming and expensive wet-lab experiments. We present the Organelx e-Science Web Server which hosts three sequence localisation predictive algorithms: In-Pero and In-Mito for classifying sub-peroxisomal and sub-mitochondrial protein localisations given their FASTA sequences, as well as the Is-PTS1 algorithm for detecting and validating potential peroxisomal proteins carrying a PTS1 signal. These tools can be used for a fast and accurate screening while looking for new peroxisomal and mitochondrial proteins. To our knowledge, this is the only service that provides these functionalities and can fasten the daily research of the peroxisomal science community.

bioinformatics↗

bioRxiv2021

In-Pero: Exploiting deep learning embeddings of protein sequences to predict the localisation of peroxisomal proteins

Peroxisomes are ubiquitous membrane-bound organelles, and aberrant localisation of peroxisomal proteins contributes to the pathogenesis of several disorders. Many computational methods focus on assigning protein sequences to subcellular compartments, but there are no specific tools tailored for the sub-localisation (matrix vs membrane) of peroxisome proteins. We present here In-Pero, a new method for predicting protein sub-peroxisomal cellular localisation. In-Pero combines standard machine learning approaches with recently proposed multi-dimensional deep-learning representations of the protein amino-acid sequence. It showed a classification accuracy above 0.9 in predicting peroxisomal matrix and membrane proteins. The method is trained and tested using a double cross-validation approach on a curated data set comprising 160 peroxisomal proteins with experimental evidence for sub-peroxisomal localisation. We further show that the proposed approach can be easily adapted (In-Mito) to the prediction of mitochondrial protein localisation obtaining performances for certain classes of proteins (matrix and inner-membrane) superior to existing tools. All data sets and codes are available at https://github.com/MarcoAnteghini and at www.systemsbiology.nl

bioinformatics↗

bioRxiv2019

Corruption of the Pearson correlation coefficient by measurement error: estimation, bias, and correction under different error models

Correlation coefficients are abundantly used in the life sciences. Their use can be limited to simple exploratory analysis or to construct association networks for visualization but they are also basic ingredients for sophisticated multivariate data analysis methods. It is therefore important to have reliable estimates for correlation coefficients. In modern life sciences, comprehensive measurement techniques are used to measure metabolites, proteins, gene-expressions and other types of data. All these measurement techniques have errors. Whereas in the old days, with simple measurements, the errors were also simple, that is not the case anymore. Errors are heterogeneous, non-constant and not independent. This hampers the quality of the estimated correlation coefficients seriously. We will discuss the different types of errors as present in modern comprehensive life science data and show with theory, simulations and real-life data how these affect the correlation coefficients. We will briefly discuss ways to improve the estimation of such coefficients.

bioinformatics↗

bioRxiv2018

Consistency, Inconsistency and Ambiguity of Metabolite Names in Biochemical Databases Used for Genome Scale Metabolic Modelling

Genome scale metabolic models (GEMs) are manually curated repositories describing the metabolic capabilities of an organism. GEMs have been successfully used in different research areas, ranging from systems medicine to biotechnology. However, the different naming conventions (namespaces) of databases used to build GEMs limit model reusability and prevent the integration of existing models. This problem is known in the GEM community but its extent has not been analyzed in depth. In this study, we investigate the name ambiguity and the multiplicity of non-systematic identifiers and we highlight the (in)consistency in their use in eleven biochemical databases of biochemical reactions and the problems that arise when mapping between different namespaces and databases. We found that such inconsistencies can be as high as 83.1%, thus emphasizing the need for strategies to deal with these issues. Currently, manual verification of the mappings appears to be the only solution to remove inconsistencies when combining models. Finally, we discuss several possible approaches to facilitate (future) unambiguous mapping.

bioinformatics↗

bioRxiv2018

Expected and observed genotype complexity in prokaryotes: correlation between 16S-rRNA phylogeny and protein domain content

BackgroundThe omnipresent 16S ribosomal RNA gene (16S-rRNA) is commonly used to identify and classify bacteria though it does not take into account the distinctive functional characteristics of taxa. We explored functional domain landscapes of over 5700 complete bacterial genomes, representing a wide coverage of the bacterial tree of life, and investigated to what extent the observed protein domain diversity correlates with the expected evolutionary diversity, using 16S-rRNA as metric for evolutionary distance.\n\nResultsAnalysis of protein domains showed that 83% of the bacterial genes code for at least one of the 9722 domain classes identified. By comparing clade specific and global persistence scores, candidate horizontal gene transfer and signifying domains could be identified. 16S-rRNA and functional domain content distances were used to evaluate and compare species divergence and overall a sigmoid curve is observed. Already at close 16S-rRNA evolutionary distances, high levels of functional diversity can be observed. At a larger 16S-rRNA distance, functional differences accumulate at a relatively lower pace.\n\nConclusionsAnalysis of 16S-rRNA sequences in the same taxa suggests that, in many cases, additional means of classification are required to obtain reliable phylogenetic relationships. Whole genome protein domain class phylogenies correlate with, and complement 16S-rRNA sequence-based phylogenies. Moreover, domain-based phylogenies can be constructed over large evolutionary distances and provide an in-depth insight of the functional diversity within and among species and enables large scale functional comparisons. The increased granularity obtained paves way for new applications to better predict the relationships between genotype, physiology and ecology.

genomics↗

bioRxiv2018

Simulation and reconstruction of metabolite-metabolite association networks using a metabolic dynamic model and correlation based-algorithms

Biological networks play a paramount role in our understanding of complex biological phenomena and metabolite-metabolite association networks are now commonly used in metabolomics applications. In this study we evaluate the performance of several network inference algorithms (PCLRC, MRNET, GENIE3, TIGRESS and modifications of the MR-NET algorithm, together with standard Pearsons and Spearmans correlation) using as a test case data generated using a dynamic metabolic model describing the metabolism of arachidonic acid (consisting of 83 metabolites and 131 reactions) and simulation individual metabolic profiles of 550 subjects. The quality of the reconstructed metabolite-metabolite association networks was assessed against the original metabolic network taking into account different degrees of association among the metabolites and different sample size and noise levels. We found that inference algorithms based on resampling and bootstrapping to perform better when correlations are used as indexes to measure the strength of metabolite-metabolite associations. We also advocate for the use of data generated using dynamic models to test the performance of algorithms for network inference since they produce correlation patterns which are more similar to those observed in real metabolomics data.

systems biology↗