arXiv subjects

Matteo Pellegrini

Publications and source records attributed to Matteo Pellegrini.

4 recordsLinked to original sources

arXiv2025

Challenges and Recommendations in Establishing National Human Diversity Genomic Projects

Genomic approaches have revolutionized medical research, providing valuable insights into human physiology and disease. Despite major benefits from large collections of genomes, the lack of diversity in genomic data represents a significant challenge for advancing biomedical discovery and accessible health solutions worldwide. Establishing a national genomic project is not a one-size-fits-all endeavor, as each country presents distinct challenges and opportunities. We identify challenges in the way of obtaining and publishing data from Whole Genome Sequencing (WGS) of people in various countries, discuss the progress made by some in their efforts to study their genetic diversity, and assess the most common issues. We recognize that a successful national genome database requires addressing several major issues, including the variable awareness of the recent developments in genomics among government officials, healthcare administrators, and policymakers, the absence of regulations, and ethical considerations, the challenges in securing funding, establishing legal frameworks, and building the necessary infrastructure. By assembling a diverse team of experts across 19 countries, we aim to provide a balanced approach in our recommendations to establish national projects. Our study acknowledges and addresses major intricacies and nuances specific to various settings and regions while presenting diverse opinions of scientists from both high-resource and low-resource countries contributing to a more inclusive and globally relevant framework for advancing genomic research and its applications.

q-bio.OT↗

arXiv2024

Embed-Search-Align: DNA Sequence Alignment using Transformer Models

DNA sequence alignment involves assigning short DNA reads to the most probable locations on an extensive reference genome. This process is crucial for various genomic analyses, including variant calling, transcriptomics, and epigenomics. Conventional methods, refined over decades, tackle this challenge in 2 steps: genome indexing followed by efficient search to locate likely positions for given reads. Building on the success of Large Language Models in encoding text into embeddings, where the distance metric captures semantic similarity, recent efforts have explored whether the same Transformer architecture can produce embeddings for DNA sequences. Such models have shown early promise in classifying short DNA sequences, such as detecting coding/non-coding regions, and enhancer, promoter sequences. However, performance at sequence classification tasks does not translate to sequence alignment, where it is necessary to search across the genome to align each read, a significantly longer-range task. We bridge this gap by framing the Sequence Alignment task for Transformer models as an "Embed-Search-Align" task. In this framework, a novel Reference-Free DNA Embedding model generates embeddings of reads and reference fragments, which are projected into a shared vector space where the read-fragment distance is used as a surrogate for alignment. Technical contributions include: (1) Contrastive loss for self-supervised training of DNA sequence representations, facilitating rich reference-free, sequence-level embeddings, and (2) a DNA vector store to enable search across fragments on a global scale. DNA-ESA is 99% accurate when aligning 250-length reads onto a human genome (3gb), rivaling conventional methods such as Bowtie and BWA-Mem. DNA-ESA exceeds the performance of 6 Transformer model baselines such as Nucleotide Transformer, Hyena-DNA, and shows task transfer across chromosomes and species.

q-bio.GN↗

arXiv2023

Challenges and perspectives in computational deconvolution of genomics data

Deciphering cell type heterogeneity is crucial for systematically understanding tissue homeostasis and its dysregulation in diseases. Computational deconvolution is an efficient approach estimating cell type abundances from a variety of omics data. Despite significant methodological progress in computational deconvolution in recent years, challenges are still outstanding. Here we enlist four significant challenges related to computational deconvolution, from the quality of the reference data, generation of ground truth data, limitations of computational methodologies, and benchmarking design and implementation. Finally, we make recommendations on reference data generation, new directions of computational methodologies and strategies to promote rigorous benchmarking.

q-bio.OT↗

arXiv2004

Position Dependent and Independent Evolutionary Models Based on Empirical Amino Acid Substitution Matrices

Evolutionary models measure the probability of amino acid substitutions occurring over different evolutionary distances. We examine various evolutionary models based on empirically derived amino acid substitution matrices. The models are constructed using the PAM and BLOSUM amino acid substitution matrices. We rescale these matrices by raising them to powers to model substitution patterns that account for different evolutionary distances. We also examine models that account for the dissimilarity of substitution rates along a protein sequence. We compare the models by computing the likelihood of each model across different alignments. We also present a specific example to illustrate the subtle differences in the estimation of evolutionary distance computed using the different models.

q-bio.PE↗