In-silico personalized protein–protein interaction networks prioritize candidate compounds for glioblastoma
5 affiliations
- 1Faculty of Mathematics and Computer Science, University of Bucharest, Bucharest, Romania
- 2Bioinformatics, National Institute of Research and Development for Biological Sciences, Bucharest, Romania
- 3Revvity Inc., Turku, Finland
- 4Faculty of Business and Administration, University of Bucharest, Bucharest, Romania
- 5Department of Mathematics and Statistics, University of Turku, Turku, Finland
Abstract
The incidence of cancer continues to rise globally, with many tumor types remaining difficult to treat effectively. While existing drugs can alleviate symptoms or slow disease progression, their efficacy varies considerably between patients. To explore this challenge, we present a computational approach for identifying patient-specific candidate compounds. Our approach was developed for glioblastoma, but it can be applied to other types of cancer. Our study involves identifying possible disease-relevant proteins specific to each patient that underlie the generated networks, together with drug targets and proteins associated with cancer dependency. These networks are analyzed using centrality measures and Louvain community detection method. Then, further investigating the individual networks, network controllability analysis is applied to identify key regulatory targets within the network. These targets are filtered and ranked to prioritize candidate drugs for individualized treatment. By comparing results across patients and against networks based on generic data, we show substantial differences between patient-specific and generic network representations. Additionally, we track changes in patient-specific networks over time in five glioblastoma cases, exploring how molecular differences between primary and recurrent tumors may influence network structure and computational drug prioritization. Author summary: Our study was motivated by the poor prognosis of glioblastoma and by the possibility that compounds not identified across all patients may still be relevant in specific molecular contexts. Therefore, our goal was to identify candidate compounds at the individual patient level using a computational network-based approach. To achieve this, we constructed a network for each sample by integrating proteins encoded by differentially expressed genes and mutated genes, along with proteins known to influence cancer progression and the targets of therapeutic drugs. These networks were based on protein-protein interactions, as they are well suited for capturing network-level effects. The study was conducted from a broad perspective to a detailed analysis. We identified hubs across all the networks, then divided the samples in communities and characterized them to understand the different ways the disease might progress. Next, we applied network controllability on each network as our main analytical method. Our in silico results suggest substantial differences between individual patient networks and a generalized network representation, indicating that patient-specific molecular context can influence computational drug prioritization. The identified compounds should be regarded as computationally prioritized candidates requiring further biological and clinical validation.
Related
- Personalized Therapeutic Strategies in Glioblastoma: Bridging Molecular Markers and In Vivo Observations for Targeted Treatment
- Targeting post-translational modifications to overcome glioblastoma therapy resistance: emerging strategies and natural compounds
- Functional divergence of the synapsin family in glioblastoma: Synapsin III as a candidate differentiation-restoring node
- Candidate compounds associated with sourness and astringency in Corni Fructus: an integrated analysis of chemical profiles, sensory evaluation, and molecular docking.
- Paired single-nucleus and spatial transcriptomics of the lung adenocarcinoma invasion sequence reveal niche domains and candidate reversal compounds
BibTeX
@article{siminea2026insilico,
title = {{In-silico personalized protein–protein interaction networks prioritize candidate compounds for glioblastoma}},
author = {Nicoleta Siminea and Victor-Bogdan Popescu and Mihaela Păun and Ion Petre and Andrei Păun},
journal = {PLOS Computational Biology},
year = {2026},
month = sep,
doi = {10.1371/journal.pcbi.1014785},
url = {https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014785},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - In-silico personalized protein–protein interaction networks prioritize candidate compounds for glioblastoma
AU - Nicoleta Siminea
AU - Victor-Bogdan Popescu
AU - Mihaela Păun
AU - Ion Petre
AU - Andrei Păun
PY - 2026
DA - 2026/09/22
JO - PLOS Computational Biology
DO - 10.1371/journal.pcbi.1014785
UR - https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014785
AB - The incidence of cancer continues to rise globally, with many tumor types remaining difficult to treat effectively. While existing drugs can alleviate symptoms or slow disease progression, their efficacy varies considerably between patients. To explore this challenge, we present a computational approach for identifying patient-specific candidate compounds. Our approach was developed for glioblastoma, but it can be applied to other types of cancer. Our study involves identifying possible disease-relevant proteins specific to each patient that underlie the generated networks, together with drug targets and proteins associated with cancer dependency. These networks are analyzed using centrality measures and Louvain community detection method. Then, further investigating the individual networks, network controllability analysis is applied to identify key regulatory targets within the network. These targets are filtered and ranked to prioritize candidate drugs for individualized treatment. By comparing results across patients and against networks based on generic data, we show substantial differences between patient-specific and generic network representations. Additionally, we track changes in patient-specific networks over time in five glioblastoma cases, exploring how molecular differences between primary and recurrent tumors may influence network structure and computational drug prioritization. Author summary: Our study was motivated by the poor prognosis of glioblastoma and by the possibility that compounds not identified across all patients may still be relevant in specific molecular contexts. Therefore, our goal was to identify candidate compounds at the individual patient level using a computational network-based approach. To achieve this, we constructed a network for each sample by integrating proteins encoded by differentially expressed genes and mutated genes, along with proteins known to influence cancer progression and the targets of therapeutic drugs. These networks were based on protein-protein interactions, as they are well suited for capturing network-level effects. The study was conducted from a broad perspective to a detailed analysis. We identified hubs across all the networks, then divided the samples in communities and characterized them to understand the different ways the disease might progress. Next, we applied network controllability on each network as our main analytical method. Our in silico results suggest substantial differences between individual patient networks and a generalized network representation, indicating that patient-specific molecular context can influence computational drug prioritization. The identified compounds should be regarded as computationally prioritized candidates requiring further biological and clinical validation.
SN - 1553-7358
ER -