Biology subjects

Silva, A. S.

Publications and source records attributed to Silva, A. S..

7 recordsLinked to original sources

bioRxiv2026

Evolutionary implications of phenotype switching for growth and survival of tumors in stochastic environments.

Tumors evolve under constantly changing microenvironmental conditions, and a key major mechanism by which cancer cells adapt to fluctuating environments is stochastic phenotypic switch via epithelial-mesenchymal plasticity (EMP). Here we demonstrate the existence of EMP in human breast epithelial cell lines and assess the fitness of E and M cells in environments that fluctuate between favorable and harsh conditions. We then develop a theoretical framework for analysis of tumor evolution with E-M stochastic phenotypic switch that integrates the behaviors of E and M cells. We characterize the outcomes of tumor evolution via three characteristic times describing i) tumor growth, ii) phenotypic adaptation and iii) fluctuations of the environment. We identify a tradeoff between tumor growth and survival, with different phenotype switching rates maximizing each of these objectives. An anticancer therapeutic strategy is proposed based on tumor survival analysis, revealing that blocking mesenchymal to epithelial transition, rather than epithelial to mesenchymal transition, is critical under fluctuating conditions. Thus, this study elucidates fundamental evolutionary mechanisms of tumor phenotypic adaptation in fluctuating environments with potentially important clinical implications.

evolutionary biology↗

bioRxiv2026

Instance-Wise Contrastive Graph Neural Network Enables the Discovery of Novel Aedes aegypti Larvicidal Compounds

Aedes aegypti remains a major arboviral vector, making larval control a critical strategy to reduce mosquito populations. However, resistance to commercial larvicides has reduced the long-term effectiveness of current interventions, reinforcing the need for new compounds with improved potency and selectivity. Here, we present an instance-wise contrastive graph neural network (GNN) framework to accelerate the discovery of novel larvicidal compounds. The model was trained on a curated dataset of 556 organic compounds organized into LC50-derived multitask classification thresholds and integrated Transformer-inspired graph learning with whole-molecule and fragment-level contrastive regularization. This model achieved strong held-out performance, with global AUC = 0.95 {+/-} 0.01, PR-AUC = 0.93 {+/-} 0.01, and MCC = 0.77 {+/-} 0.03, outperforming conventional machine learning and graph-based baselines. Predictive uncertainty analysis and counterfactual maps further supported the interpretation of threshold-sensitive predictions and substructural contribution patterns. The model was applied to screen 1.3 million compounds, resulting in 10 candidates for experimental validation. Three compounds showed measurable larvicidal activity against A. aegypti larvae. Among them, LC-79 emerged as the most promising hit, with 2-day and 5-day LC50 values of 0.24 {micro}g/mL (0.66 {micro}M) and 0.05 {micro}g/mL (0.13 {micro}M), respectively, an IE50 of 0.06 {micro}g/mL (0.16 {micro}M), and rapid larval mortality (LT50 = 1.10 days at 1 {micro}g/mL). LC-79 also showed no measurable acute toxicity to Daphnia magna at the highest tested concentration [EC50-48h >43 {micro}g/mL (>119 {micro}M)], resulting in selectivity indices >180 and >860 relative to its 2-day and 5-day LC50 values. Overall, this study demonstrates that contrastive graph learning can move beyond retrospective larvicide modeling to experimentally validated hit discovery, identifying LC-79 as a potent and preliminarily selective acylthiourea larvicide candidate for further mechanism-of-action, resistance, and semi-field evaluation.

bioinformatics↗

bioRxiv2025

The biomolecular profiles of extracellular vesicles from odontogenic stem cell lines depict donor-dependent differences and emphasize their therapeutic and regenerative potential

BackgroundStem cell-derived extracellular vesicles (EVs) hold great promise in regenerative medicine. However, a comprehensive understanding of the regenerative capabilities of EVs from different stem cell sources remains limited. MethodsThis study systematically compares EVs derived from three odontogenic cell types. Analyses includes EV isolation and characterization, cell viability assays, vasculogenesis experiments, proteomic profiling, and miRNA sequencing. ResultsAll three EV types displayed similar surface marker profiles. Dental pulp stem cell-derived EVs showed superior cellular uptake, promoted higher cell proliferation, and enhanced vasculogenesis compared to periodontal ligament stem cell-derived EVs. Gingival fibroblast-derived EVs performed similarly in functional assays. Principal component analysis of miRNA profiles revealed strong biological heterogeneity among EV sources, with donor-specific factors exerting a greater influence on EV characteristics than cellular origin--an aspect underexplored in prior studies. ConclusionsThese findings underscore the complexity of EV functionality and highlight the regenerative potential of dental stem cell-derived EVs.

cell biology↗

bioRxiv2025

Support for the deuterostome clade comes from systematic errors

There is a long-standing consensus that the animal phyla closest to our own phylum of Chordata are the Echinodermata and Hemichordata. These three phyla constitute the major clade of Deuterostomia. Recent analyses have questioned the support for the monophyly of Deuterostomia, however, showing that the branch leading to deuterostomes is very short and may be influenced by systematic error. Here we use a site-by-site approach to explore multiple sources of error. Under conditions that promote long-branch attraction (LBA) - especially branch-length heterogeneity and sites constrained in their amino acid composition - we find that deuterostome monophyly is strongly supported. When we make efforts to mitigate these sources of error, we cannot distinguish between monophyletic and paraphyletic Deuterostomia. Our findings have implications for the interpretation of putative deuterostome fossils, for the reconstruction of a bilaterian ancestor and, more generally, for how datasets for deep-time phylogenetic analyses are assembled and analyzed. TeaserThe apparently close relationship between Chordata and Ambulacraria (echinoderms and hemichordates) is boosted by a long-branch attraction artefact.

evolutionary biology↗

bioRxiv2022

Replicative Instability Drives Cancer Progression

In the past decade, defective DNA repair has been increasingly linked with cancer progression. Human tumors with markers of defective DNA repair and increased replication stress have been shown to exhibit genomic instability and poor survival rates across tumor types. Here we utilize-omics data from two independent consortia to identify the genetic underpinnings of replication stress, therapy resistance, and primary carcinoma to brain metastasis in BRCA wildtype tumors. In doing so, we have defined a new pan-cancer class of tumors characterized by replicative instability (RIN). RIN is defined by genomic evolution secondary to replicative challenge. Our data supports a model whereby defective single-strand break repair, translesion synthesis, and non-homologous end joining effectors drive RIN. Collectively, we find that RIN accelerates cancer progression by driving copy number alterations and transcriptional program rewiring that promote tumor evolution. Statement of SignificanceDefining the genetic basis of genomic instability with wildtype BRCA repair effectors is a significant unmet need in cancer research. Here we identify and characterize a pan-cancer cohort of tumors driven by replicative instability (RIN). We find that RIN drives therapy resistance and distant metastases across multiple tumor types.

cancer biology↗

bioRxiv2020

An Advanced Framework for Time-lapse Microscopy Image Analysis

Time-lapse microscopy is a powerful technique that generates large volumes of image-based information to quantify the behaviors of cell populations. This method has been applied to cancer studies to estimate the drug response for precision medicine and has great potential to address inter-patient (or intertumoral) heterogeneity. A couple of algorithms exist to analyze time-lapse microscopy images; however, most deal with very high-resolution images involving few cells (typically cell lines). There are currently no advanced and efficient computational frameworks available to process large-scale time-lapse microscopy imaging data to estimate patient-specific response to therapy based on a large population of primary cells. In this paper, we propose a robust and user-friendly pipeline to preprocess the images and track the behaviors of thousands of cancer cells simultaneously for a better drug response prediction of cancer patients. Availability and ImplementationSource code is available at: https://github.com/CompbioLabUCF/CellTrack ACM Reference FormatQibing Jiang, Praneeth Sudalagunta, Mark B. Meads, Khandakar Tanvir Ahmed, Tara Rutkowski, Ken Shain, Ariosto S. Silva, and Wei Zhang. 2020. An Advanced Framework for Time-lapse Microscopy Image Analysis. In Proceedings of BioKDD: 19th International Workshop on Data Mining In Bioinformatics (BioKDD). ACM, New York, NY, USA, 8 pages. https://doi.org/10.1145/nnnnnnn.nnnnnnn

bioinformatics↗

bioRxiv2019

Training load and body composition in adults practicing cyclical exercises

Although meta-analyzes point to a weight loss of no more than 3 kg to exercise, body fat of the athletes are below of the population. Then training load may be a determining factor in body composition. This study verified if dose of physical training adopted by exercise practitioners is determinant in body composition. Was a cross-sectional retrospective study carried out with 122 individuals (45.8 {+/-} 13.0 years, 50 men) who practiced cyclic exercises (running, walking or cycling) randomly recruited in six regions which the city was geographically divided. Caloric expenditure was estimated in the trainings based on the frequency, intensity and duration of the exercises and the body composition was assessed by electrical bioimpedance. The subjects practiced 4.3 {+/-} 1.5 weekly sessions, with mean duration of 56.7 {+/-} 28.2 minutes/session and caloric expenditure/day of 410.2 {+/-} 384.1 kcal/day. Linear regression test revealed a negative correlation (p=0.000) between the mean daily expenditure and all measures of adiposity tested (absolute and relative body fat and visceral fat), and evidenced that the training load explains 56% of the proposed model. When adjusted for sex, the correlation remained in men and disappeared in women. Mens with energy expenditure higher than 785 kcal/day presented lower fat stores than congeners with minor diary training load. Conclude that training load adopted by physical exercise practitioners is an influencing factor in the body composition of men, but not of women. Load adopted in conventional programs training seems insufficient to produce adequate body composition.

physiology↗