Michele Ceccarelli

This study identifies how epigenetic changes and spatial clustering influence immune dynamics and therapy resistance in melanoma through single-cell and spatial transcriptomics analyses.

This study shows GABA receptor B signaling promotes GBM growth in females through immune mechanisms, with sex-specific effects on survival and treatment outcomes.

This study identifies ADAR as a key mediator of immune evasion in H3K27M+ pediatric gliomas and shows that ATRA can induce ADAR degradation, enhancing antiviral responses and tumor immunogenicity.

This study shows that DPP-4 inhibition improves CD8+ T cell metabolism and anti-tumor activity in preclinical and clinical settings.

This study evaluates CNN and ViT models for predicting gene expression from spatial transcriptomics images, finding that compact CNNs with larger patches perform better for biologically relevant genes.

Fate-AI improves cancer detection sensitivity and generalization by integrating fragmentomic and methylation features from cfDNA sequencing.

The paper identifies DNA methylation subtypes in metastatic melanoma and links them to clinical outcomes and immune response, suggesting methylome targeting for precision immunotherapy.

This study shows targeting ZNF638 induces viral mimicry and enhances immune checkpoint inhibition in glioblastoma by increasing dsRNA signaling and PD-L1 expression.

This study shows DNA methylation patterns classify pleural mesothelioma cells into immune profiles, and DHA treatment alters these profiles to improve immune response.

The paper evaluates deconvolution methods for inferring immune infiltration from bulk gene expression using a DREAM Challenge, finding that some community-contributed methods, especially deep learning-based ones, perform well.

This study shows that oncometabolite D-2HG drives enhancer hypermethylation in IDH-mutant gliomas, leading to hyporesponsive microglia and reduced immune reactivity.

This study identifies a panel of 20 microRNAs as potential diagnostic biomarkers for breast cancer using feature selection methods and network analysis.

This paper introduces a biologically informed neural network model for predicting drug sensitivity and synergy using multi-omics data, showing improved performance on imbalanced datasets.

The paper develops a deep learning classifier to distinguish somatic and germline mutations in tumor-only samples, achieving high accuracy across tissue types and reducing racial bias in TMB estimates.

SCEVAN is a fast variational algorithm that deconvolves clonal substructure in tumors from single-cell data, accurately distinguishing malignant and non-malignant cells.

BayesDeBulk estimates immune and stromal cell composition from bulk proteomic and transcriptomic data, outperforming existing tools in synthetic and real data examples.

This study predicts non-coding RNA functions from raw sequence data using deep learning, achieving good accuracy without requiring secondary structure computation.

The paper introduces scTHI, a method to identify ligand-receptor interactions in glioma using single-cell RNA sequencing data from six human datasets.

TCGAbiolinksGUI provides a graphical interface for analyzing GDC cancer data, making the TCGAbiolinks package more accessible to non-R users.

RGBM identifies transcriptional regulators of glioma subtypes by inferring gene regulatory networks from diverse data sources and outperforming existing methods on benchmark datasets.

This paper introduces DCD, a method to detect differential sub-networks in paired biological networks, showing it outperforms existing methods in simulated and real cancer datasets.

The paper introduces an improved method for detecting statistically significant network changes in biological networks, showing better performance on dense random geometric networks and application to glioma cancer regulatory networks.