Search

In the past years, Graph Neural Networks (GNNs) have become the `de facto' standard in various deep learning domains, thanks to their flexibility in modeling real-world phenomena represented as graphs. However, the message-passing mechanism of GNNs faces challenges in learnability and expressivity, hindering high performance on heterophilic graphs, where adjacent nodes frequently have different labels. Most existing solutions addressing these challenges are primarily confined to specific benchmarks focused on node classification tasks. This narrow focus restricts the potential impact that link prediction under heterophily could offer in several applications, including recommender systems. For example, in social networks, two users may be connected for some latent reason, making it challenging to predict such connections in advance. Physics-Inspired GNNs such as GRAFF provided a significant contribution to enhance node classification performance under heterophily, thanks to the adoption of physics biases in the message-passing. Drawing inspiration from these findings, we advocate that the methodology employed by GRAFF can improve link prediction performance as well. To further explore this hypothesis, we introduce GRAFF-LP, an extension of GRAFF to link prediction. We evaluate its efficacy within a recent collection of heterophilic graphs, establishing a new benchmark for link prediction under heterophily. Our approach surpasses previous methods, in most of the datasets, showcasing a strong flexibility in different contexts, and achieving relative AUROC improvements of up to 26.7%., Comment: 7 pages, 1 figure

The artist similarity quest has become a crucial subject in social and scientific contexts. Modern research solutions facilitate music discovery according to user tastes. However, defining similarity among artists may involve several aspects, even related to a subjective perspective, and it often affects a recommendation. This paper presents GATSY, a recommendation system built upon graph attention networks and driven by a clusterized embedding of artists. The proposed framework takes advantage of a graph topology of the input data to achieve outstanding performance results without relying heavily on hand-crafted features. This flexibility allows us to introduce fictitious artists in a music dataset, create bridges to previously unrelated artists, and get recommendations conditioned by possibly heterogeneous sources. Experimental results prove the effectiveness of the proposed method with respect to state-of-the-art solutions., Comment: 6 pages, Submitted to MLSP 2023

Robust biomarkers of aging have been developed from DNA methylation in humans and more recently, in mice. This study aimed to generate a novel epigenetic clock in rats-a model with unique physical, physiological, and biochemical advantages-by incorporating behavioral data, unsupervised machine learning, and network analysis to identify epigenetic signals that not only track with age, but also relates to phenotypic aging. Reduced representation bisulfite sequencing (RRBS) data was used to train an epigenetic age (DNAmAge) measure in Fischer 344 CDF (F344) rats. This measure correlated with age at (r = 0.93) in an independent sample, and related to physical functioning (p=5.9e-3), after adjusting for age and cell counts. DNAmAge was also found to correlate with age in male C57BL/6 mice (r = 0.79), and was decreased in response to caloric restriction. Our signatures driven by CpGs in intergenic regions that showed substantial overlap with H3K9me3, H3K27me3, and E2F1 transcriptional factor binding.

Robust biomarkers of aging have been developed from DNA methylation in humans and more recently, in mice. This study aimed to generate a novel epigenetic clock in rats-a model with unique physical, physiological, and biochemical advantages-by incorporating behavioral data, unsupervised machine learning, and network analysis to identify epigenetic signals that not only track with age, but also relates to phenotypic aging. Reduced representation bisulfite sequencing (RRBS) data was used to train an epigenetic age (DNAmAge) measure in Fischer 344 CDF (F344) rats. This measure correlated with age at (r = 0.93) in an independent sample, and related to physical functioning (p=5.9e-3), after adjusting for age and cell counts. DNAmAge was also found to correlate with age in male C57BL/6 mice (r = 0.79), and was decreased in response to caloric restriction. Our signatures driven by CpGs in intergenic regions that showed substantial overlap with H3K9me3, H3K27me3, and E2F1 transcriptional factor binding.

Robust biomarkers of aging have been developed from DNA methylation in humans and more recently, in mice. This study aimed to generate a novel epigenetic clock in rats-a model with unique physical, physiological, and biochemical advantages-by incorporating behavioral data, unsupervised machine learning, and network analysis to identify epigenetic signals that not only track with age, but also relates to phenotypic aging. Reduced representation bisulfite sequencing (RRBS) data was used to train an epigenetic age (DNAmAge) measure in Fischer 344 CDF (F344) rats. This measure correlated with age at (r = 0.93) in an independent sample, and related to physical functioning (p=5.9e-3), after adjusting for age and cell counts. DNAmAge was also found to correlate with age in male C57BL/6 mice (r = 0.79), and was decreased in response to caloric restriction. Our signatures driven by CpGs in intergenic regions that showed substantial overlap with H3K9me3, H3K27me3, and E2F1 transcriptional factor binding.

Background: Heart Transplantation (HTX) is the only curative treatment available for patients with end-stage heart failure (HF).During the first year post-transplantation more than 25% of patients will go through rejection episodes and will face the risk of developing rejection with consequent graft dysfunction with an increased morbidity and mortality. Preventing and treating acute rejection is the most central task for clinicians working with transplanted patients. The ISHLT 2005 and 2013 working formulations defined the histopathologic profile of three types of rejection: Cellular (ACR) Humoral (AMR) and Mixed (MIX). Nowadays serial endomyocardial biopsies (EMB) at decreasing intervals during the first year after transplantation and laboratory tests, such as Donor Specific Antibody (DSA) measurements, remain the gold-standard in diagnosing and monitoring acute rejection but they are morbid and prone to artefacts of sampling, interpretation and testing methodologies. Therefore this histopathological assessment needs integrative new biomarkers to characterize risk stratification for outcomes in heart transplantation. To date, the exact mechanisms involved in rejection after solid transplantation are not completely understood, so investigating process that contribute to acute allograft rejection and find effective biomarkers to diagnose, monitoring and predicting rejection will be of great value for the development of improved anti-rejection strategies. The advent of sequencing technology such as Next Generation Sequencing (NGS) is changing medical genomics by accelerating new disease biomarkers discovery. MicroRNAs (miRNAs) are small non-coding RNA molecules (19-24 nucleotides), highly conserved, which regulate genes expression at the post transcriptional level. Aim: The aim of this study is to identify MicroRNA (miRNAs) expression profile in the first year after heart transplantation (HTX) with Next Generation Sequencing (NGS) technology in formalin fixed paraffin