Welcome to my website!
I'm Cristiano Cordì, a bioinformatician with a passion for machine learning and regulatory genomics.
Here you can find information about my projects, CV, and how to contact me.
Projects
TRIDENT
Tri-Dimensional Embeddings Navigation Tool for interactive data visualization.
Master Thesis
Deciphering organism-wide gene regulation with genomic deep learning in C. elegans.
OpenNano
GeoMX Spatial Transcriptomics Data Analysis platform.
Bachelor Thesis
CCRL2 expression combined to specialized endothelial cell signature in human lung cancers.
TRIDENT - Tri-Dimensional Embeddings Navigation Tool
View on GitHub
Overview
TRIDENT is a Blender add-on designed for bioinformaticians who need to generate three-dimensional visualizations of dimensionality reduction methods such as UMAP, t-SNE, or PCA.
By using the rendering power of the Blender EEVEE engine, TRIDENT provides realtime navigation of data clouds, seamless color labeling of categories, and the ability to export publication-ready figures or smooth animations for presentations.
Key Features
- Real-time 3D visualization of embeddings
- Interactive navigation and customization controls
- Support for the visualization of multiple embedding algorithms
- Export capabilities for publication-ready figures
Technologies Used
Python, C++, Blender
OpenNano - GeoMX Spatial Transcriptomics Analysis
Overview
Open Nano is a free open-source python package developed to address a lack of Python-based tools for analyzing Nanostring GeoMx DSP data.
Documentation
The package comes with comprehensive documentation that can be found on the OpenNano website.
Key Features
- Automated data preprocessing pipelines
- Spatial expression visualization
- Statistical analysis tools
- Integration with existing workflows
Technologies Used
Python, NumPy, SciPy, Matplotlib, AnnData, Scanpy, DESeq2
Master Thesis - Genomic Deep Learning in C. elegans
Abstract
Understanding how the genome controls gene expression to create distinct cell types is a central problem in biology. DNA encodes regulatory instructions, but the mechanisms remain unclear.
Here, I've applied deep learning to decode this regulatory grammar and predict single-cell gene expression. Using convolutional and attention-based models on Caenorhabditis elegans sequences, I've leveraged its well-mapped cellular landscape to study expression across all cell types. The models predict expression directly from DNA and represent a major advance.
I've also built an integrated dataset spanning developmental stages and applied explainable machine learning to reveal regulatory features. This work offers new insight into genome regulation with applications in development and disease.
Key Achievements
- Developed novel neural network architectures for genomic sequence analysis
- Identified previously unknown regulatory motifs
- Created comprehensive gene regulation maps
- Published findings in peer-reviewed journal
Technologies Used
Python, TensorFlow, Keras, CREsted, CUDA, HPC
View Thesis Document
Bachelor Thesis - CCRL2 in Lung Cancer Prognosis
Abstract
Investigation of CCRL2 expression patterns combined with specialized endothelial cell signatures as prognostic markers in human lung cancer patients.
Key Findings
- Identified CCRL2 as significant prognostic marker
- Characterized endothelial cell signatures
- Developed predictive models for patient outcomes
- Validated results across multiple patient cohorts
Technologies Used
R, Bioconductor, Survival Analysis, Statistics
Resume
Professional Summary
I am a bioinformatician with expertise in machine learning, regulatory genomics, and single-cell transcriptomics.
Education
- Master's Degree in Bioinformatics - Katholieke Universiteit Leuven
- Bachelor's Degree in Bioinformatics - La Sapienza - University of Rome
Key Skills
- Machine Learning & Deep Learning
- Genomics & Transcriptomics Analysis
- Python, R, Java, Bash, TensorFlow, Keras
- High Performance Computing (HPC)
- Scientific Writing & Research
Research Areas
- Regulatory Genomics
- Single-cell Analysis
- Spatial Transcriptomics
- Computational Biology
- Data Visualization
Projects
I've developed open-source tools for the bioinformatics community, OpenNano and TRIDENT.
