Biology subjects

Costa, R. S.

Publications and source records attributed to Costa, R. S..

6 recordsLinked to original sources

bioRxiv2026

On why and how to encode probability distributions on graph representations of omics data: enhancing predictive tasks and knowledge discovery

The growing availability and complexity of omics data have driven the development of specialized algorithms for modeling molecular systems. Although graph-based learning methods effectively represent biological interactions, they often neglect the statistical information embedded in node and edge annotations. To address this limitation, we propose a novel graph-based framework that integrates structured statistical distributions into nodes and edges, capturing probabilistic characteristics of molecular relationships. We evaluate the proposed approach on omics datasets from five cancer types across multiple clinical outcomes, including survivability and primary tumor site. Results demonstrate predictive performance comparable to established machine learning baselines. Beyond prediction, the statistically enriched graph representations enable the identification and characterization of regulatory modules associated with clinical outcomes, enhancing biological interpretability. These findings suggest that incorporating structured statistical information into graph representations provides a competitive and interpretable framework for predictive modeling and knowledge discovery in complex diseases.

bioinformatics↗

bioRxiv2023

Pattern-centric transformation of omics-data sources grounded on multi-wise gene associations aids predictive tasks in TCGA while ensuring interpretability.

MotivationThe increasing prevalence of omics data sources is pushing the study of regulatory mechanisms underlying complex diseases such as cancer. However, the vast quantities of features produced and the inherent interplay between them lead to a level of complexity that hampers both descriptive and predictive tasks, requiring custom-built algorithms that can extract relevant information from these sources of data. ResultsWe propose a transformation that moves data centered on molecules (e.g. transcripts and proteins) to a new data space focused on putative regulatory modules given by statistically relevant patterns of coexpression. The proposed transformation extracts patterns from the data through biclustering and uses them to create new variables with guarantees of interpretability and discriminative power. The transformation is shown to achieve dimensionality reductions of up to 99% and to increase the predictive performance of various classifiers across multiple omics layers. Our results suggest that a transformation of omics data from gene-centric to pattern-centric data provides benefits to both prediction tasks and human interpretation. The proposed approach is expected to greatly support further bioinformatic analyses for precision medicine applications. AvailabilitySoftware code and the raw results generated are available at github.com/Andrempp/Pattern-Centric-Transformation. Contactandremppatricio@tecnico.ulisboa.pt Supplementary informationSupplementary data are available at Journal Name online.

genomics↗

bioRxiv2022

A general deep hybrid model for bioreactor systems: combining first Principles equations with deep neural networks

Numerous studies have reported the use of hybrid semiparametric systems that combine shallow neural networks with mechanistic models for bioprocess modeling. Here we revisit the general bioreactor hybrid modeling problem and introduce some of the most recent deep learning techniques. The single layer networks were extended to multi-layer networks with varying depths and combined with First Principles equations in the form of deep hybrid models. Deep learning techniques, namely the adaptive moment estimation method (ADAM), stochastic regularization and depth-dependent weights initialization were evaluated. Modified sensitivity equations are proposed for the computation of gradients in order to reduce CPU time for the training of deep hybrid models. The methods are illustrated with applications to a synthetic dataset and a pilot 50 L MUT+ Pichia pastoris process expressing a single chain antibody fragment. All in all, the results point to a systematic generalization improvement of deep hybrid models over its shallow counterpart. Moreover, the CPU cost to train the deep hybrid models is shown to be lower than for the shallow counterpart. In the pilot 50L MUT+ Pichia pastoris data set, the prediction accuracy was increased by 18.4% and the CPU decreased by 43.4%. HighlightsO_LIShallow hybrid models have been widely used for bioprocess modeling and optimization C_LIO_LINon-deep training using e.g. the Levenberg - Marquardt method, cross-validation and indirect sensitivity equations have been the methods of choice C_LIO_LIDeep learning with ADAM, stochastic regularization and indirect sensitivity significantly reduces the training CPU C_LIO_LIThe generalization capacity of deep hybrid models systematically outperforms that of shallow hybrid models C_LI

systems biology↗

bioRxiv2021

Enrichment analysis on regulatory subspaces: a novel direction for the superior description of cellular responses to SARS-CoV-2

StatementThe enrichment analysis of discriminative cell transcriptional responses to SARS-CoV-2 infection using biclustering produces a broader set of superiorly enriched GO terms and KEGG pathways against alternative state-of-the-art machine learning approaches, unraveling novel knowledge. Motivation and methodsThe comprehensive understanding of the impacts of the SARS-CoV-2 virus on infected cells is still incomplete. This work identifies and analyses the main cell regulatory processes affected and induced by SARS-CoV-2, using transcriptomic data from several infectable cell lines available in public databases and in vivo samples. We propose a new class of statistical models to handle three major challenges, namely the scarcity of observations, the high dimensionality of the data, and the complexity of the interactions between genes. Additionally, we analyse the function of these genes and their interactions within cells to compare them to ones affected by IAV (H1N1), RSV and HPIV3 in the target cell lines. ResultsGathered results show that, although clustering and predictive algorithms aid classic functional enrichment analysis, recent pattern-based biclustering algorithms significantly improve the number and quality of the detected biological processes. Additionally, a comparative analysis of these processes is performed to identify potential pathophysiological characteristics of COVID-19. These are further compared to those identified by other authors for the same virus as well as related ones such as SARS-CoV-1. This approach is particularly relevant due to a lack of other works utilizing more complex machine learning tools within this context.

bioinformatics↗

bioRxiv2021

DISA tool: discriminative and informative subspace assessment with categorical and numerical outcomes

MotivationPattern discovery and subspace clustering play a central role in the biological domain, supporting for instance putative regulatory module discovery from omic data for both descriptive and predictive ends. In the presence of target variables (e.g. phenotypes), regulatory patterns should further satisfy delineate discriminative power properties, well-established in the presence of categorical outcomes, yet largely disregarded for numerical outcomes, such as risk profiles and quantitative phenotypes. ResultsDISA (Discriminative and Informative Subspace Assessment), a Python software package, is proposed to assess patterns in the presence of numerical outcomes using well-established measures together with a novel principle able to statistically assess the correlation gain of the subspace against the overall space. Results confirm the possibility to soundly extend discriminative criteria towards numerical outcomes without the drawbacks well-associated with discretization procedures. A case study is provided to show the properties of the proposed method. AvailabilityDISA is freely available at https://github.com/JupitersMight/DISA under the MIT license. Contact{leonardoalexandre@tecnico.ulisboa.pt,rmch@tecnico.ulisboa.pt} and {rs.costa@fct.unl.pt}

bioinformatics↗

bioRxiv2020

EMBRYONIC DEVELOPMENT OF THE FIRE-EYE-TETRA Moenkhausia oligolepis (CHARACIFORMES: CHARACIDAE)

This study describes the embryonic development of Moenkhausia oligolepis in captive conditions. After fertilization, the embryos were collected every 10 min up to 2 h, every 20 min up to 4 h, and every 30 min until hatching. The fertilized eggs of M. oligolepis measured approximately 0.85 {+/-} 0.5 mm and have an adhesive surface. The embryonic development lasted 14 hours at 25{degrees}C, with the Zygote, Cleavage, Blastula, Gastrula, Neurula and Segmentation phases. The hatching occurred in embryos around the 30-somites stage. Our results bring only the second description of embryonic development to a species of Moenkhausia genus, the first for the refereed species. Such data are of paramount importance considering the current conflicting state of this genus phylogenetic classification and may help taxonomic studies. Understand the biology of a species that is easily handling in captive conditions and has an ornamental appeal may assist studies in its reproduction in order to both, supply the aquarium market and help the species conservation in nature. Moreover, our data enable the M. oligolepis to be used as a model species in biotechnological applications, such germ cell transplantation approach.

developmental biology↗