Biology subjects

Guccione, C.

Publications and source records attributed to Guccione, C..

5 recordsLinked to original sources

bioRxiv2025

Community assembly modeling of microbial evolution within Barrett's esophagus and esophageal adenocarcinoma

Mathematical modeling of somatic evolution, a process impacting both host cells and microbial communities in the human body, can capture important dynamics driving carcinogenesis. Here we considered models for esophageal adenocarcinoma (EAC), a cancer that has dramatically increased in incidence over the past few decades in Western populations, with high case fatality rates due to late-stage diagnoses. Despite advancements in genomic analyses of the precursor Barretts esophagus (BE), prevention of late-stage EAC remains a significant clinical challenge. Previous microbiome studies in BE and EAC have focused on quantifying static microbial abundance differences rather than evolutionary dynamics. Using whole genome sequencing data from esophageal tissues, we first applied a robust bioinformatics pipeline to extract non-host DNA reads, mapped these putative reads to microbial taxa, and retained those taxa with high genomic coverage. When applying mathematical models of microbial evolution to sequential stages of progression to EAC, we observed evidence of neutral dynamics in community assembly within normal esophageal tissue and BE, but not EAC. In a case-control study of BE patients who progressed to EAC cancer outcomes (CO) versus those who had non-cancer outcomes (NCO) during follow-up (mean=10.5 years), we found that Helicobacter pylori deviated significantly from the neutral expectation in BE NCO, suggesting that factors related to H. pylori or H. pylori infection itself may influence EAC risk. Additionally, simulations incorporating selection recapitulated non-neutral behaviors observed in the datasets. Formally modeling dynamics during progression holds promise in clinical applications by offering a deeper understanding of microbial involvement in cancer development.

cancer biology↗

bioRxiv2024

Predicting locus phylogenetic utility using machine learning

Disentangling evolutionary signal from noise in genomic datasets is essential to building phylogenies. The efficiency of current sequencing platforms and workflows has resulted in a plethora of large-scale phylogenomic datasets where, if signal is weak, it can be easily overwhelmed with non-phylogenetic signal and noise. However, the nature of the latter is not well understood. Although certain factors have been investigated and verified as impacting the accuracy of phylogenetic reconstructions, many others (as well as interactions among different factors) remain understudied. Here we use a large simulation-based dataset and machine learning to better understand the factors, and their interactions, that contribute to species tree error. We trained Random Forest regression models on the features extracted from simulated alignments under known phylogenies to predict the phylogenetic utility of the loci. Loci with the worst utility were then filtered out, resulting in an improved signal-to-noise ratio across the dataset. We investigated the relative importance of different features used by the model, as well as how they correspond to the originally simulated properties. We further used the model on several diverse empirical datasets to predict and subset the least reliable loci and re-infer the phylogenies. We measure the impacts of the subsetting on the overall topologies, difficult nodes identified in the original studies, as well as branch length distribution. Our results suggest that subsetting based on the utility predicted by the model can improve the topological accuracy of the trees and their average statistical support, and limits paralogy and its effects. Although the topology generated from the filtered datasets may not always be dramatically different from that generated from unfiltered data, the worst loci consistently yielded different topologies and worst statistical support, indicating that our protocol identified phylogenetic noise in the empirical data.

evolutionary biology↗

bioRxiv2023

BIRDMAn: A Bayesian differential abundance framework that enables robust inference of host-microbe associations

Quantifying the differential abundance (DA) of specific taxa among experimental groups in microbiome studies is challenging due to data characteristics (e.g., compositionality, sparsity) and specific study designs (e.g., repeated measures, meta-analysis, cross-over). Here we present BIRDMAn (Bayesian Inferential Regression for Differential Microbiome Analysis), a flexible DA method that can account for microbiome data characteristics and diverse experimental designs. Simulations show that BIRDMAn models are robust to uneven sequencing depth and provide a >20-fold improvement in statistical power over existing methods. We then use BIRDMAn to identify antibiotic-mediated perturbations undetected by other DA methods due to subject-level heterogeneity. Finally, we demonstrate how BIRDMAn can construct state-of-the-art cancer-type classifiers using The Cancer Genome Atlas (TCGA) dataset, with substantial accuracy improvements over random forests and existing DA tools across multiple sequencing centers. Collectively, BIRDMAn extracts more informative biological signals while accounting for study-specific experimental conditions than existing approaches.

bioinformatics↗

bioRxiv2022

AI-assisted Discovery of an Ethnicity-influenced Driver of Cell Transformation in Esophageal and Gastroesophageal Junction Adenocarcinomas

Although Barretts metaplasia of the esophagus (BE) is the only known precursor lesion to esophageal adenocarcinomas (EACs), drivers of the metaplasia[->]dysplasia[->]neoplasia cascade in the esophagus remains incompletely understood. Using an AI-guided network transcriptomics approach, in which EAC initiation and progression is modeled as networks to simplify complex multi-cellular processes, we first predict cellular continuum states and disease driving processes with an unprecedented degree of precision. Key AI-guided predictions are subsequently validated in a human organoid model and patient-derived biopsies of BE, a case-control study of genomics of BE progression, and in a cross-sectional study of 113 patients with BE and EACs. We find that all EACs must originate from BE, pinpoint a CXCL8/IL8{leftrightarrow}neutrophil immune microenvironment as a driver of cellular transformation in both EACs and gastroesophageal junction-ACs. This driver is prominent in Caucasians (Cau), but notably absent in African Americans (AAs). Network-derived gene signatures, independent signatures of neutrophil processes, CXCL8/IL8, and an absolute neutrophil count (ANC) are associated with risk of progression. SNPs associated with ethnic changes in ANC modify that risk. Thus, findings define a racially influenced immunological basis for cell transformation and suggest that benign ethnic neutropenia in AAs may serve as a deterrent to BE[->]EAC progression. BRIEF SUMMARYEsophageal adenocarcinoma (EAC) is a highly lethal cancer among Caucasians, while African Americans are somewhat protected; what factors drive transformation with racial disparity remain unknown. AI-enabled creation of the first computational map of neoplastic progression in the esophagus built and validated using transcriptomic datasets from diverse cohorts of human samples pinpointed CXCL8{leftrightarrow}neutrophil tumor immune-microenvironment as a racially influenced driver of EACs and GEJ-ACs. Computational tools pinpoint a racially influenced driver of cell transformation during BE[->]EAC progression; in doing so, it reveals new novel biology, informs disease modeling, therapeutic strategies, and biomarkers. LAY SUMMARYBy modeling diseases as networks, this work unravels a fundamental race-influenced immunologic driver of cell transformation in adenocarcinomas of the esophagus and the gastroesophageal junction.

cancer biology↗

bioRxiv2019

All-FIT: Allele-Frequency-based Imputation of Tumor Purity from High-Depth Sequencing Data

MotivationClinical sequencing aims to identify somatic mutations in cancer cells for accurate diagnosis and treatment. However, most widely used clinical assays lack patient-matched control DNA and additional analysis is needed to distinguish somatic and unfiltered germline variants. Such computational analyses require accurate assessment of tumor cell content in individual specimens. Histological estimates often do not corroborate with results from computational methods that are primarily designed for normal-tumor matched data and can be confounded by genomic heterogeneity and presence of sub-clonal mutations.\n\nMethodsAll-FIT is an iterative weighted least square method to estimate specimen tumor purity based on the allele frequencies of variants detected in high-depth, targeted, clinical sequencing data.\n\nResultsUsing simulated and clinical data, we demonstrate All-FITs accuracy and improved performance against leading computational approaches, highlighting the importance of interpreting purity estimates based on expected biology of tumors.\n\nAvailability and ImplementationFreely available at http://software.khiabanian-lab.org.

bioinformatics↗