Alexandre Rossi Paschoal

The paper compares chromosome-scale genomes of two B. caapi ethnovarieties to identify genomic features associated with ethnovariety differentiation.

PanTEon provides a cross-kingdom framework for reproducible TE classification, benchmarking machine learning methods and highlighting performance variations across TE superfamilies and lineages.

The paper evaluates feature representations for predicting cancer-associated sORFs in non-coding RNA using traditional machine learning models and finds that k-mer features outperform genomic language model embeddings.

This paper introduces DeepSEA, a deep learning tool that functionally annotates antimicrobial resistance proteins without sequence alignment, achieving higher recall than alignment-based methods.

The paper introduces a protocol for creating and implementing gene dictionaries across annotated genomes using a Python CLI and Jupyter Notebooks, producing FAIR-compliant .gdict files.

This study introduces GENNUS, a generative approach using GANs and SMOTE to improve ncRNA classification by addressing class imbalance through synthetic data generation.

CODARFE predicts continuous environmental variables from microbiome data, outperforming existing tools in correlation and prediction accuracy across multiple datasets.

This study classifies LTR retrotransposons at the superfamily level using Predictive Bi-Clustering Trees, showing comparable performance to existing methods in D. melanogaster and A. thaliana.

This paper presents the first telomere-to-telomere genome of Theobroma grandiflorum, revealing genomic structure, evolutionary relationships, and gene variations that contribute to its unique traits and domestication.

The paper analyzes the genome and population genomics of Coffea arabica to reveal its diversification history and breeding-related genetic changes.

This study evaluates the performance of lncRNA prediction tools on long-read sequencing data and identifies challenges in their application.

This paper compares mathematical models for feature extraction from biological sequences, showing their effectiveness in classifying Long Non-Coding RNA and other sequence types.

TERL classifies transposable elements using CNNs, achieving high accuracy and speed across multiple datasets and classification levels.

mirtronDB is a database that compiles and organizes mirtron data from 18 species, allowing users to search and browse mirtron precursors and mature sequences.