Skip to main content

EDITORIAL article

Front. Appl. Math. Stat., 21 March 2023
Sec. Mathematical Biology
This article is part of the Research Topic Data science in neuro- and onco-biology View all 6 articles

Editorial: Data science in neuro- and onco-biology

  • 1Mida, Department of Mathematics, Università di Genova, Genova, Italy
  • 2Faculty of Medicine and Health Technology and BioMediTech Institute, Tampere University, Tampere, Finland
  • 3LISCOMP, IRCCS Ospedale Policlinico San Martino, Genova, Italy

Editorial on the Research Topic
Data science in neuro- and onco-biology

In many fields of biomedical research the development and use of breakthrough technologies often produce a huge amount of heterogeneous, multi-scale data ranging from molecular genomic data to images describing the function and structure of tissues [1, 2]. In contrast, at a clinical level the need for real-time, minimally-invasive, and cost-effective measurements often reduces the quality of the recorded data rendering them to be noisy and incomplete [3].

Data science approaches based on robust mathematical and statistical methods have proven themselves to be of crucial importance in both these scenarios. For example, machine learning and data mining approaches can be used for patients classification and modeling of disease progression [4, 5]. Similarly, computational models and simulations may be used to build a patient's digital twin opening the possibilities for highly personalized therapeutic plans [6].

The purpose of this Research Topic is to review and present recent mathematical and statistical developments in the analysis of biomedical data. To this end, we focused on two specific biomedical fields, namely neuroscience and oncology, motivated by the fact that cancer and various neurodegenerative dementias, e.g., due to Alzheimer's and Parkinson's diseases, are among the leading causes of death worldwide1.

The Research Topic features two Reviews and three Original Research articles that are briefly summarized below.

Ioannides et al. reviewed two families of numerical approaches for dimensionality reduction of complex systems, based on principal component analysis and graph clustering theory, respectively. The reviewed methods are illustrated on two neuroscientific studies: (i) the study of correlations between sleep stages and brain activity; (ii) the characterization of brain response to median nerve stimulation. In both scenarios, magnetoencephalography and electroencephalography were used for monitoring brain activity. To prove the flexibility of the reviewed methods, the authors also considered a problem rising from astrophysics, namely clustering of ultraluminous infrared galaxies based on their spectral energy distribution.

Righetti et al. revised recent mechanistic models developed for modeling at a molecular-scale the homeostasis of α-synuclein, one of the key drivers of Parkinson's disease parthenogenesis [7]. Specifically they grouped the considered models in two families: single-pathway chemical kinetic model of α-synuclein aggregation, and multiple-pathway models accounting for both α-synuclein aggregation and degradation. Possible applications of such models toward quantitative system pharmacologic are discussed.

Reconstructing high quality magnetic resonance images is a first crucial step in many modern quantitative approaches for cancer precision medicine, including radiogenomics [8]. Motivated by this consideration, Di Cola et al. proposed to combine Rudin-Osher-Fatemi (ROF) total variation minimization with an histogram-matching (HM) approach for magnetic resonance image denoising. Different combinations of various implementations of both the ROF denoising algorithm and the HM approach were tested and validated on simulated data extracted from a publicly available database.

Nieus et al. present a multi-class logistic regression algorithm with l1 penalization to infer sparse connectivity networks within large populations of neurons from their voltage tracers. The authors proved the robustness of the proposed methods with respect to different parameters (such as presence of noise, network size, and limited data availability) by simulating realistic networks comprising both excitatory and inhibitory neurons. Overall, this work is motivated by the spread of emerging techniques capable of simultaneously recording spikes and post-synaptic-potential from large populations of neurons. Such techniques comprise genetically encoded voltage indicators [9] or proper combination of different optical recording techniques [10].

Sapienza et al. performed an experimental study on the impact of parameter initialization on the performance and training time of a modern autoencoder architecture, called deep image prior (DIP) [11]. Specifically, they focused on the application of DIP for denoising x-ray (sparse) computed tomography (CT) images. First, they used natural RGB images and gray-scaled CT images of a phantom to select a set of best initial configuration. Then, the selected initial configurations were applied for artifact removal on the CT images of a coronavirus (COVID-19) patient's lungs.

Author contributions

All authors contributed to the article and approved the submitted version.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

References

1. Alber M, Buganza Tepole A, Cannon WR, De S, Dura-Bernal S, Garikipati K, et al. Integrating machine learning and multiscale modeling perspectives, challenges, and opportunities in the biological, biomedical, and behavioral sciences. NPJ Digit Med. (2019) 2:115. doi: 10.1038/s41746-019-0193-y

PubMed Abstract | CrossRef Full Text | Google Scholar

2. Phan JH, Quo CF, Cheng C, Wang MD. Multiscale integration of-omic, imaging, and clinical data in biomedical informatics. IEEE Rev Biomed Eng. (2012) 5:74–87. doi: 10.1109/RBME.2012.2212427

PubMed Abstract | CrossRef Full Text | Google Scholar

3. Subbiah V. The next generation of evidence-based medicine. Nat Med. (2023) 29:49–58. doi: 10.1038/s41591-022-02160-z

PubMed Abstract | CrossRef Full Text | Google Scholar

4. Glaab E, Rauschenberger A, Banzi R, Gerardi C, Garcia P, Demotes J. Biomarker discovery studies for patient stratification using machine learning analysis of omics data: a scoping review. BMJ Open. (2021) 11:e053674. doi: 10.1136/bmjopen-2021-053674

PubMed Abstract | CrossRef Full Text | Google Scholar

5. Myszczynska MA, Ojamies PN, Lacoste AM, Neil D, Saffari A, Mead R, et al. Applications of machine learning to diagnosis and treatment of neurodegenerative diseases. Nat Rev Neurol. (2020) 16:440–56. doi: 10.1038/s41582-020-0377-8

PubMed Abstract | CrossRef Full Text | Google Scholar

6. Hernandez-Boussard T, Macklin P, Greenspan EJ, Gryshuk AL, Stahlberg E, Syeda-Mahmood T, et al. Digital twins for predictive oncology will be a paradigm shift for precision cancer care. Nat Med. (2021) 27:2065–6. doi: 10.1038/s41591-021-01558-5

PubMed Abstract | CrossRef Full Text | Google Scholar

7. Poewe W, Seppi K, Tanner CM, Halliday GM, Brundin P, Volkmann J, et al. Parkinson disease. Nat Rev Dis Primers. (2017) 3:1–21. doi: 10.1038/nrdp.2017.13

PubMed Abstract | CrossRef Full Text | Google Scholar

8. Lo Gullo R, Daimiel I, Morris EA, Pinker K. Combining molecular and imaging metrics in cancer: radiogenomics. Insights Imaging. (2020) 11:1–17. doi: 10.1186/s13244-019-0795-6

PubMed Abstract | CrossRef Full Text | Google Scholar

9. Gong Y, Huang C, Li JZ, Grewe BF, Zhang Y, Eismann S, et al. High-speed recording of neural spikes in awake mice and flies with a fluorescent voltage sensor. Science. (2015) 350:1361–6. doi: 10.1126/science.aab0810

PubMed Abstract | CrossRef Full Text | Google Scholar

10. Brondi M, Bruzzone M, Lodovichi C, Dal Maschio M. Optogenetic methods to investigate brain alterations in preclinical models. Cells. (2022) 11:1848. doi: 10.3390/cells11111848

PubMed Abstract | CrossRef Full Text | Google Scholar

11. Ulyanov D, Vedaldi A, Lempitsky V. Deep image prior. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Salt Lake City (2018). p. 9446–54.

Google Scholar

Keywords: machine learning, biomedical data analysis, oncobiology, neurobiology, system biology, complex system, imaging

Citation: Sommariva S, Subramaniyam NP and Piana M (2023) Editorial: Data science in neuro- and onco-biology. Front. Appl. Math. Stat. 9:1175331. doi: 10.3389/fams.2023.1175331

Received: 27 February 2023; Accepted: 10 March 2023;
Published: 21 March 2023.

Edited and reviewed by: Raluca Eftimie, University of Franche-Comté, France

Copyright © 2023 Sommariva, Subramaniyam and Piana. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Sara Sommariva, c29tbWFyaXZhJiN4MDAwNDA7ZGltYS51bmlnZS5pdA==

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.