Claudio Lorenzi

👋 Hi, I’m Claudio Lorenzi ( @CloXD ): Bioinformatician, full stack software developer interested in Machine Learning and Artificial Intelligence.

Who am I

Committed and enthusiastic bioinformatician with over six years of experience in development of bioinformatics tools and analysis of biological data. Conﬁdent in working with diﬀerent programing languages and frameworks. Fast learner and always curious about new technologies, especially in the machine learning ﬁeld. Collaborative and creative, promoter of an healthy and productive working environment.

Skills

Programming Languages

C++
Python
R
JavaScript/TypeScript
Java

Frameworks and tools

Git
Singularity and Docker
Nextflow
VSCode and Eclipse
Several softwares and R libraries for Bioinformatics

Data Analysis

RNA-seq, ChIP-seq, Genomics
Statistical tests, pathway analysis, basics of protein structure analysis

Machine Learning

SVM, Linear and Logistic regression, Naive Bayes, Decision tree based models, k-NN
Deep learning, CNN, Reinfrocement Learning, GAN
Clustering techniques ( hierarchical, k-means, EM, DBSCAN, SOM)
Feature reduction ( PCA, t-SNE, NMF, Autoencoders)

Web/GUI Development

Angular
Node.js (Express.js and Sequelize)
mySQL
RESTful API
Electron
Bootstrap and Material design

Languages

Italian
English
French

Works

PCaProfiler: the Prostate Cancer atlas Profiler, the web tool that enables you to mine clinical and transcriptional data of over 1365 clinical samples across various stages. Associations with prostate cancer progression can be quantified using a quantitative pseudotime progression score. In addition, the tool also allows the annotation of new experimental and clinical data for disease progression.
iMOKA: a tool to identify k-mers as classifing features from large datasets of sequencing data. The software is mainly written in C++ ( MLPack as ML library ) and Pyhton ( scikit-learn and TF libraries ) for the core part and Angular + Electron for the GUI.
IRFinder-S: a new version of the popular software for the identification and quantification of intron retention events. Includes a CNN based filter that discerns artefactual IR events using visual-like features ( TensorFlow Python library) and a IR database/visualization web tool, IRbase
MAGeCK_View: a small web page that allows to visualize the results of CRISPR screening generated by MAGeCK.
Cell2Patients: I contributed to my colleague’s PhD project that aim to transfer the knowledge aquired from cell lines model to patient data
PickPocket: an attempt to create an automatic protein pocket classifier that uses secondary structure informations alongside with other useful features. Unfortunately, our tests didn’t show a real improvement respect to other existing methods, but it was a good attempt.

Working Experience

Bioinformatician - PostDoc @IOR (Institute of Oncology Research, Bellinzona - CH)
Bioinformatician - PhD Candidate @IGH ( Institute of Human Genetics, Montpellier - FR )
Study and Development Engineer @IMGT (the international ImMunoGeneTics infromation system, Montpellier - FR)

Eductation

PhD in Bioinformatics - University of Montpellier
M.Sc in Bioinformatics - University of Bologna
B.Sc in Medical Biotechnology - University of Milan