Computational Health @Unipi
The computational health group applies computing technologies to enhance biomedical research and to improve health and quality of life. AI technologies as well as bioinformatics and computational biology are used to have an impact on knowledge extraction and integration, predictive biomarker identification, mechanism of action elucidation and dynamic quantitative modeling of biological systems. The applicative domains are medicine, pharmacology, biotechnology and nutrition science.
Members
Paolo Ferragina
Full Professor
Alina Sîrbu
Associate Professor
Corrado Priami
Full Professor
Silvia Giulia Galfrè
PostDoc
Giuseppe Prencipe
Associate Professor
Marco Podda
PostDoc
Projects
MEDICA: Modelling and vErification of alkaptonuria and multiple sclerosis Driven by biomedICAl data PRIN 2022 Project, 2023. @misc{medica, Abstract: The goal of this project is to bridge the gap between data, models and validation, integrating the data-driven and model-driven research directions, so that to develop next generation modelling and analysis approaches for biomedical investigations. We are tackling open research questions for Multiple Sclerosis (MS) and Alkaptonuria (AKU), for which we have world-leading experts in our consortium. |
I-POTERI: Innovation in the telemonitoring of the rehabilitation of post-operative cardiac patients Collaboration with the IFC CNR Pisa and several SME Tuscan companies, 2021. (Abstract) @misc{I-POTERI, Abstract: Cardiac rehabilitation in the post-surgical phase has an important impact on the patient as it speeds up his psychophysical recovery, and improves the clinical picture by reducing the sense of fatigue, dyspnea, and increases survival. In this context, compliance with the rehabilitation program is a key element for the therapeutic benefits and, at the same time, for providing elements for strengthening the patient, in terms of increasing awareness of their own state of health. To achieve these results, the I-POTERI project aims to evolve current telemonitoring systems by developing and industrializing a SW-HW platform based on the integration of home kits (consisting of wearable and low-invasive medical devices), on the some monitoring microservices, and on innovative AI technology, which will allow the continuous monitoring of numerous parameters both during the patient's normal home activities and during rehabilitation activities. The ultimate goal is to customize and optimize the post-surgical rehabilitation process and to allow personalized intervention by the doctor given the parameters of the rehabilitation process and any anomalies of the monitored vital parameters. The project also aims to carry out a retrospective analysis in terms of the effectiveness of the performance and adequacy of the program, thanks also to the application of data analytics algorithms and artificial intelligence techniques. |
MOTIF-RAPTOR Collab. with Massachusetts General Hospital, Harvard Medical School, Broad Institute of MIT and Harvard, 2020. (Abstract) @misc{MOTIF-RAPTOR, This project aims at designing a computational tool which hinges on Transcription Factors (TF) and integrates sequence-based predictive models, chromatin accessibility, gene expression datasets and GWAS summary statistics to systematically investigate how TF function is affected by genetic variants. We have successfully tested Motif-Raptor on complex traits such as rheumatoid arthritis and red blood cell count and demonstrated its ability to prioritize relevant cell types, potential regulatory TFs and non-coding SNPs which have been previously characterized and validated. |
PANS Collaboration with Stanford Medical School, 2020. (Abstract) @misc{PANS, This project aims to uncover the mechanisms involved in the development and remission of symptoms for PANS (pediatric acute-onset neuropsychiatric syndrome). It involves analysis of proteomic and metabolomic data for a set of PANS patients at different disease stages and under different treatments. |
Activity and Sleep monitoring Collaboration with Biobeats and UNIPI Pharmacy Department, 2020. (Abstract) @misc{sleep_monitoring, This project explores the interaction between physical and psychological characteristics of healthy individuals, by monitoring various parameters in a set of volunteers, leading to a multi-level analysis. The parameters include accelerometer data, sleep quality, hormone levels, heart rate, stress, emotions, and others, and can provide a wide view of how different processes interact with each other during daily life. |
PROMET (SPARK Project) Collaboration with the Pisa University Hospital Oncology Unit , 2020. (Abstract) @misc{PROMET, Circulating tumor DNA can be detected in cancer patients after tumor resection. This project aims to evaluate whether sequencing of circulating tumor DNA can be employed in prognosis for cancer patients, concentrating on patients with colorectal cancer liver metastasis. This consists of both survival prediction and selection of prognosis biomarkers. |
COVID-19 Collaboration with the Pisa University Hospital and CNR Pisa, 2020. (Abstract) @misc{covid19, Clinical data for COVID-19 patients can be employed to predict disease progression and final outcomes. However, the quantity of clinical variables and missing data may pose challenges. We are developing a method to select the most important clinical variables and to predict mortality at hospitalisation time. |
DROP (SPARK project) Collaboration with the University of Pisa Medical School, 2020. (Abstract) @misc{DROP, While whole slide imaging is providing detailed information to pathologists, diagnosis and therapeutic decisions are still mostly based on expert evaluation of the digital images, which is subjective introducing errors and tampering with reproducibility. This project aims to develop a prototype tool that can aid the evaluation of biomarkers in breast tumour tissues, which employs image analysis and machine learning. |
Vaccination coverage prediction for Meningitis B Collaboration with GlaxoSmithKlein, 2020. (Abstract) @misc{meningitisB, The project aims to identify genomic characteristics of Meningococcus B strains that predict whether an existing vaccine is able to protect against new isolates of the bacterium. While the state of the art uses pure statistical tools to predict efficiency, this project aims to employ machine learning tools based on Deep Learning. The hope is that these methods, being based on complete genetic sequences, will be more generic and translatable to other vaccines. |
Protein selection for vaccine development Collaboration with GlaxoSmithKlein, 2020. (Abstract) @misc{vaccine, Vaccine development typically includes a preliminary in silico selection of proteins that can be promising candidates to be included in the vaccine. Current approaches start by annotating proteins with various parameters and then selecting based on the annotations. This project aims to eliminate the intermediate phase of annotating the sequences, by considering the protein sequences directly as input to a classifier. |
Covid-19 morbidity and mortality in type 2 diabetes mellitus: possible roles of macrophages and of glucose lowering drugs Collaboration with University of Parma, 2020. (Abstract) @misc{covid-diabetes, The Severe Acute Respiratory Syndrome-Coronavirus-2 (SARS-CoV-2) causes COVID-19, which, if severe, features interstitial pneumonia, multiple organ damage, macrophage (MΦ) activation syndrome and vascular thrombosis. People with type 2 diabetes (T2DM) may be more vulnerable to severe COVID-19. AIMS: 1. To assess the independent impact of T2DM on the risk of death or transfer to intensive care unit (ICU) of patients admitted to the Covid-1 macro-unit of Ospedale Maggiore. 2. To assess, in a SARS-CoV-dependent model of inflamed human MΦ: a. the role of T2DM and of some anti-diabetes agents (ADA) in modulating MΦ inflammation; b. the effects of the MΦ secretome on lung microvascular endothelium. |
Multicriteria Data Structures and Algorithms: from compressed to learned indexes, and beyond. PRIN no. 2017WR7SHH, 2019. @misc{PRIN2017, The ever growing need to efficiently store, retrieve and analyze massive datasets, is currently made more complex by the different requirements posed by users, devices and applications. Such a new level of complexity cannot be handled properly by current data structures for big data problems. This project lays down the theoretical and algorithmic-engineering foundations of a new generation of Multicriteria Data Structures and Algorithms that will seamlessly integrate, via a principled optimization approach, modern compressed data structures with new, revolutionary, data structures learned from the input data by using proper machine-learning tools, with applications to key-value stores and bioinformatics tools. |
Selected Publications
2024
Podda, Marco; Bonechi, Simone; Palladino, Andrea; Scaramuzzino, Mattia; Brozzi, Alessandro; Roma, Guglielmo; Muzzi, Alessandro; Priami, Corrado; Sîrbu, Alina; Bodini, Margherita
Classification of Neisseria meningitidis genomes with a bag-of-words approach and machine learning Journal Article
In: iScience, 2024.
@article{podda2024classification,
title = {Classification of Neisseria meningitidis genomes with a bag-of-words approach and machine learning},
author = {Marco Podda and Simone Bonechi and Andrea Palladino and Mattia Scaramuzzino and Alessandro Brozzi and Guglielmo Roma and Alessandro Muzzi and Corrado Priami and Alina Sîrbu and Margherita Bodini},
url = {https://www.cell.com/iscience/pdf/S2589-0042(24)00478-4.pdf},
year = {2024},
date = {2024-02-16},
journal = {iScience},
publisher = {Elsevier},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Cignoni, Giacomo; Scatena, Cristian; Frascarelli, Chiara; Fusco, Nicola; Naccarato, Antonio Giuseppe; Fanelli, Giuseppe Nicoló; Sîrbu, Alina
PD-L1 Classification of Weakly-Labeled Whole Slide Images of Breast Cancer Journal Article
In: arXiv preprint arXiv:2404.10175, 2024.
@article{cignoni2024pd,
title = {PD-L1 Classification of Weakly-Labeled Whole Slide Images of Breast Cancer},
author = {Giacomo Cignoni and Cristian Scatena and Chiara Frascarelli and Nicola Fusco and Antonio Giuseppe Naccarato and Giuseppe Nicoló Fanelli and Alina Sîrbu},
url = {https://arxiv.org/abs/2404.10175},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {arXiv preprint arXiv:2404.10175},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Pingitore, Alessandro; Zhang, Chenxiang; Vassalle, Cristina; Ferragina, Paolo; Landi, Patrizia; Mastorci, Francesca; Sicari, Rosa; Tommasi, Alessandro; Zavattari, Cesare; Prencipe, Giuseppe; others,
Machine learning to identify a composite indicator to predict cardiac death in ischemic heart disease Journal Article
In: International Journal of Cardiology, vol. 404, pp. 131981, 2024.
@article{pingitore2024machine,
title = {Machine learning to identify a composite indicator to predict cardiac death in ischemic heart disease},
author = {Alessandro Pingitore and Chenxiang Zhang and Cristina Vassalle and Paolo Ferragina and Patrizia Landi and Francesca Mastorci and Rosa Sicari and Alessandro Tommasi and Cesare Zavattari and Giuseppe Prencipe and others},
url = {https://www.sciencedirect.com/science/article/pii/S016752732400531X},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {International Journal of Cardiology},
volume = {404},
pages = {131981},
publisher = {Elsevier},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Tolloso, Matteo; Galfrè, Silvia Giulia; Pavone, Arianna; Podda, Marco; S^ırbu, Alina; Priami, Corrado
How Much Do DNA and Protein Deep Embeddings Preserve Biological Information? Proceedings Article
In: International Conference on Computational Methods in Systems Biology, pp. 209–225, Springer 2024.
@inproceedings{tolloso2024much,
title = {How Much Do DNA and Protein Deep Embeddings Preserve Biological Information?},
author = {Matteo Tolloso and Silvia Giulia Galfrè and Arianna Pavone and Marco Podda and Alina S^ırbu and Corrado Priami},
url = {https://link.springer.com/chapter/10.1007/978-3-031-71671-3_15},
doi = {https://doi.org/10.1007/978-3-031-71671-3_15},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
booktitle = {International Conference on Computational Methods in Systems Biology},
pages = {209–225},
organization = {Springer},
series = {Lecture Notes in Computer Science},
abstract = {Models that map DNA and protein sequences into deep embeddings have been recently developed. While their ability to improve prediction in downstream tasks has been demonstrated, clear advantages and disadvantages of embedding types, and different means of applying them, are not yet available. In this paper we compare five different models (one for DNA, four for proteins) and different embedding aggregation methods with respect to their ability to preserve evolutionary and functional information, using a hierarchical tree approach. Specifically, we introduce a novel procedure that builds hierarchical clustering trees to assess the relative position of sequences in the embedding latent space, compared to the phylogenetic and functional similarities between sequences. The methods are benchmarked on five different datasets from various organisms. The ESM protein language model and DNABert emerge as best performers in different settings.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
2023
Testa, Irene; Prencipe, Giuseppe; Priami, Corrado; Sîrbu, Alina
Comparison of Machine Learning Classifiers on Integrated Transcriptomic Data Proceedings Article
In: 2023 IEEE International Conference on Big Data (BigData), pp. 4987–4996, IEEE 2023.
@inproceedings{testa2023comparison,
title = {Comparison of Machine Learning Classifiers on Integrated Transcriptomic Data},
author = {Irene Testa and Giuseppe Prencipe and Corrado Priami and Alina Sîrbu},
url = {https://ieeexplore.ieee.org/abstract/document/10386445},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 IEEE International Conference on Big Data (BigData)},
pages = {4987–4996},
organization = {IEEE},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
2021
Sîrbu, Alina; Barbieri, Greta; Faita, Francesco; Ferragina, Paolo; Gargani, Luna; Ghiadoni, Lorenzo; Priami, Corrado
Early outcome detection for COVID-19 patients Journal Article
In: Scientific Reports, vol. 11, no. 1, pp. 18464, 2021.
@article{sirbu2021early,
title = {Early outcome detection for COVID-19 patients},
author = {Alina Sîrbu and Greta Barbieri and Francesco Faita and Paolo Ferragina and Luna Gargani and Lorenzo Ghiadoni and Corrado Priami},
url = {https://www.nature.com/articles/s41598-021-97990-1},
year = {2021},
date = {2021-01-01},
urldate = {2021-01-01},
journal = {Scientific Reports},
volume = {11},
number = {1},
pages = {18464},
publisher = {Nature Publishing Group UK London},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
2020
Muscolino, Alessandro; Maria, Antonio Di; Alaimo, Salvatore; Borzì, Stefano; Ferragina, Paolo; Ferro, Alfredo; Pulvirenti, Alfredo
NETME: On-the-fly knowledge network construction from biomedical literature Book Section
In: International Conference on Complex Networks and their Applications (COMPLEX), Springer International Publishing, 2020.
@incollection{Muscolino2020,
title = {NETME: On-the-fly knowledge network construction from biomedical literature},
author = {Alessandro Muscolino and Antonio Di Maria and Salvatore Alaimo and Stefano Borzì and Paolo Ferragina and Alfredo Ferro and Alfredo Pulvirenti},
year = {2020},
date = {2020-01-01},
booktitle = {International Conference on Complex Networks and their Applications (COMPLEX)},
publisher = {Springer International Publishing},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
Ferragina, Paolo; Vinciguerra, Giorgio
The PGM-index: a fully-dynamic compressed learned index with provable worst-case bounds Journal Article
In: Proceedings of the VLDB Endowment, vol. 13, no. 10, pp. 1162–1175, 2020.
@article{Ferragina2020,
title = {The PGM-index: a fully-dynamic compressed learned index with provable worst-case bounds},
author = {Paolo Ferragina and Giorgio Vinciguerra},
doi = {10.14778/3389133.3389135},
year = {2020},
date = {2020-01-01},
journal = {Proceedings of the VLDB Endowment},
volume = {13},
number = {10},
pages = {1162--1175},
publisher = {VLDB Endowment},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Rossi, Alessio; Pozzo, Eleonora Da; Menicagli, Dario; Tremolanti, Chiara; Priami, Corrado; Sirbu, Alina; Clifton, David A; Martini, Claudia; Morelli, Davide
A Public Dataset of 24-h Multi-Levels Psycho-Physiological Responses in Young Healthy Adults Journal Article
In: Data, vol. 5, no. 4, pp. 91, 2020.
@article{Rossi2020,
title = {A Public Dataset of 24-h Multi-Levels Psycho-Physiological Responses in Young Healthy Adults},
author = {Alessio Rossi and Eleonora Da Pozzo and Dario Menicagli and Chiara Tremolanti and Corrado Priami and Alina Sirbu and David A Clifton and Claudia Martini and Davide Morelli},
doi = {10.3390/data5040091},
year = {2020},
date = {2020-01-01},
journal = {Data},
volume = {5},
number = {4},
pages = {91},
publisher = {MDPI AG},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
2019
Misselbeck, Karla; Parolo, Silvia; Lorenzini, Francesca; Savoca, Valeria; Leonardelli, Lorena; Bora, Pranami; Morine, Melissa J; Mione, Maria Caterina; Domenici, Enrico; Priami, Corrado
A network-based approach to identify deregulated pathways and drug effects in metabolic syndrome Journal Article
In: Nature Communications, vol. 10, no. 1, 2019.
@article{Misselbeck2019,
title = {A network-based approach to identify deregulated pathways and drug effects in metabolic syndrome},
author = {Karla Misselbeck and Silvia Parolo and Francesca Lorenzini and Valeria Savoca and Lorena Leonardelli and Pranami Bora and Melissa J Morine and Maria Caterina Mione and Enrico Domenici and Corrado Priami},
doi = {10.1038/s41467-019-13208-z},
year = {2019},
date = {2019-11-01},
journal = {Nature Communications},
volume = {10},
number = {1},
publisher = {Springer Science and Business Media LLC},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Simoni, Giulia; Vo, Hong Thanh; Priami, Corrado; Marchetti, Luca
A comparison of deterministic and stochastic approaches for sensitivity analysis in computational systems biology Journal Article
In: Briefings in Bioinformatics, vol. 21, no. 2, pp. 527–540, 2019.
@article{Simoni2019,
title = {A comparison of deterministic and stochastic approaches for sensitivity analysis in computational systems biology},
author = {Giulia Simoni and Hong Thanh Vo and Corrado Priami and Luca Marchetti},
doi = {10.1093/bib/bbz014},
year = {2019},
date = {2019-01-01},
journal = {Briefings in Bioinformatics},
volume = {21},
number = {2},
pages = {527--540},
publisher = {Oxford University Press (OUP)},
keywords = {},
pubstate = {published},
tppubtype = {article}
}