computational-oncology

Nextflow pipeline for demultiplexing ATAC and WGS reads from the DATAC single-cell co-assay, with ArchR-based quality control reporting.

Falsification-first research on cancer genetic vulnerabilities that replicate across independent CRISPR screens.

Personal profile & technical portfolio for Luca Celebrano (@Lukecele): medical student, bioinformatics and open source

Official Research Repository for TEMPO: A Novel Hybrid Neuromorphic Framework for Efficient and Interpretable Integration of Multi-Omic High-Dimensional Oncology Data

Multi-scale bioinformatics pipeline for patient-specific drug screening via computational simulation of patient-derived organoids. Integrates U-Net segmentation, agent-based cell simulation, QSAR drug modelling and PDE diffusion.

In-silico agent-based framework for coexistence-control of pancreatic cancer (PDAC) via food-medicine-homology compounds

Multi-cohort clinical survival validation of microvascular decoupling in EGFR-mutant lung adenocarcinoma (N = 854, p < 10^-15)

GNN optimization for Protein Interaction (PPI) analysis in Oncology. Implements normalized graph convolution for stable Edge AI execution on SBC

10x Visium HD Subcellular Spatial Architecture and Microvascular Decoupling in Osimertinib Resistance (λ ≈ 624 µm)

From Subcellular Spatial Biophysics to Routine Digital Pathology AI: Resolving Non-Genetic Osimertinib Resistance via Microvascular Decoupling

Multi-omic prediction of therapy-related secondary malignancy risk in solid tumour survivors, integrating mutational signatures, transcriptomics, and CNV across 1,374 TCGA patients.

Operator-theoretic spatial biology framework for tumor–immune interface analysis using Hodge decomposition, antisymmetric commutator geometry, and graph operator dynamics. Validated across HCC, IMC, CABO/NIVO, and a pan-cancer Visium cohort (26 sections, 118k+ spots, 6 cancer types).

Automated molecular classification of brain gliomas (LGG vs. GBM) using Machine Learning and clinical-mutational profiles from TCGA/UCI

Digital patient generation and drug response prediction via TCGA-DepMap integration — CVAE virtual patients, multi-level similarity scoring, 578-compound drug profiling

Edge-GNN: Constraint-aware graph neural networks for biological interaction modeling under edge deployment constraints (computational oncology, PPI networks).

LLM-driven multi-agent simulation of the KRAS G12D pancreatic tumor microenvironment — emergent immunosuppression, NK/DC agents, ablation studies. bioRxiv 2026.

Two-drug tumour resistance simulator (ODE + RK4) with a virtual clinical trial engine: Kaplan-Meier survival curves and log-rank tests over synthetic patient cohorts.

Synthetic biology Verilog logic gates for smart contextual tumor microenvironment destruction via Cello compilation flow

CT-informed biomechanical simulation of mesothelioma tumour growth with the finite element method (DolfinX, GMSH).

PERSIST (PERsister-State In-Silico Targeting): residual-state modelling of single-cell drug response for combination therapy prediction in EGFR-mutant NSCLC and TNBC