Advancing understanding of long COVID pathophysiology through quantum walk-based network analysis, 2026, Park et al.

[SNT Gatchaman]

Advancing understanding of long COVID pathophysiology through quantum walk-based network analysis

Spoiler: Authors

Park, Jaesub; Hwang, Woochang; Lee, Seokjun; Lee, Hyun Chang; MacMahon, Méabh; Zilbauer, Matthias; Han, Namshik

MOTIVATION
Long COVID is a multisystem condition characterized by persistent symptoms such as fatigue, cognitive impairment, and systemic inflammation following COVID-19 infection. However, its mechanisms remain poorly understood. In this study, we applied the quantum walk, a computational approach leveraging quantum interference, to explore large-scale SARS-CoV-2–induced protein networks.

RESULT
Compared to the conventional random walk with restart method, the quantum walk demonstrated superior capacity to traverse deeper regions of the network, uncovering proteins and pathways implicated in Long COVID. Key findings include mitochondrial dysfunction, thromboinflammatory responses, and neuronal inflammation as central mechanisms. Quantum walk uniquely identified the CDGSH iron-sulfur domain-containing protein family and VDAC1, a mitochondrial calcium transporter, as critical regulators of these processes. VDAC1 emerged as a potential biomarker and therapeutic target, supported by FDA-approved compounds such as cannabidiol. These findings highlight quantum walk as a powerful tool for elucidating complex biological systems and identifying novel therapeutic targets for conditions like Long COVID.

Web | DOI | PDF | Bioinformatics Advances | Open Access

 

[Murph]

A better title would be:
Advancing understanding of quantum walk-based network analysis through playing around with long covid.

 

[SNT Gatchaman]

Cambridge / South Korea team.

1 Cambridge Stem Cell Institute, University of Cambridge
2 Milner Therapeutics Institute, University of Cambridge
3 Cambridge Centre for AI in Medicine, Department of Applied Mathematics and Theoretical Physics, University of Cambridge
4 CardiaTec Biosciences Ltd, Cambridge, CB2 1GE, United Kingdom
5 Department of Paediatrics, University of Cambridge
6 Department of Paediatric Gastroenterology, Hepatology and Nutrition, Cambridge University Hospitals (CUH), Addenbrooke’s
7 Department of Quantum Information, Institute for Convergence Research and Education in Advanced Technology and Engineering, Yonsei University, Seoul
8 Department of Nano Biomedical Engineering (NanoBME), Advanced Science Institute, Yonsei University, Seoul
9 Center for Nanomedicine, Institute for Basic Science (IBS), Seoul

Intro selected quotes —

[LC …] underscores the need for integrative computational frameworks that combine molecular, clinical, and patient-reported data to reveal hidden biological network patterns

In recent years, quantum algorithms, including QW, have shown great potential in efficiently navigating graph-based systems, offering insights that are challenging to obtain through classical methods. Quantum walk (QW) algorithms extend the concept of classical random walks by leveraging quantum mechanics principles such as superposition and interference […] makes QW especially suited for u ering patterns in large, interconnected networks.

More recently, QW has also been applied to address a range of complex biological problems. For instance, predicting missing protein–protein interactions in incomplete PPI networks has been explored using hybrid approaches that combine continuous-time classical and quantum walks

In this study, we employed quantum-inspired approaches to analyze protein interaction networks associated with Long COVID, focusing on their ability to uncover critical biological mechanisms.

Conclusion —

By enabling deeper exploration of biological networks, QW has identified novel proteins and pathways that are critically involved in Long COVID pathophysiology. In particular, its ability to prioritize mitochondrial dysfunction and systemic processes further establishes its value in studying multisystem disorders or other post-acute sequelae

 

[SNT Gatchaman]

Methods —

Differentially expressed SomaScan measurements in 6-month Long COVID patients versus recovered patients during acute COVID-19 were obtained from Cervia-Hasler et al. (2024). A total of 1335 proteins were extracted by combining Data S1 and S3, available as supplementary data at Bioinformatics Advances online and used as differentially expressed proteins (DEPs). The proteins that were significantly up- or down-regulated (two-tailed t tests, P < .05, |log2FC| > 0) were selected.

See Persistent complement dysregulation with signs of thromboinflammation in active Long Covid (2024, Science)

The SIP [SARS-CoV-2–induced protein] network was constructed using a human protein–protein interaction (PPI) backbone sourced from the STRING database v11.5

 

[SNT Gatchaman]

Results (describing more of the approach) —

To elucidate the mechanisms underlying Long COVID, we investigated how SARS-CoV-2 impacts host proteins and pathways using a SARS-CoV-2–induced protein (SIP) network (Han et al. 2021, Hwang and Han 2022). The rationale for constructing the SIP network lies in the assumption, based on cause and effect, that proteins significantly affected during acute COVID-19 could contribute to the long-term effects

SIP network was constructed by integrating directly interacting proteins (DIPs), which directly interact with the SARSCoV-2 protein, and differentially expressed proteins (DEPs), the latter identified from proteins exhibiting differential expression during acute COVID-19.

the SIP network was designed with three distinct layers: the DIP layer, the DEP layer, and an intermediate hidden layer that bridges the two, consisting of 11 247 nodes and 179 554 unique edges.

To assess the impact of SARS-CoV-2 on the SIP network, we applied network propagation methods, QW and RWR. These methods allowed us to model how the viral signal influences proteins across the network. For further analysis, we categorized the network nodes into shallow (distance ≤ 2) and deep (distance > 2) regions, where distance refers to the distance from SARS-CoV-2, enabling us to explore the distinct roles of proximal and distal proteins. Our analysis revealed distinct patterns in the distribution of proteins across the SIP network

This enrichment of DEPs, which we know are changing in response to COVID-19 infection, in deeper regions suggests that proteins further from the SARS-CoV-2 signal play an important role in mediating the broader effects of the virus. These findings highlight the need to investigate both shallow and deep regions of the SIP network to fully capture the direct and indirect impacts of SARS-CoV-2.

Proteins that are highly connected and involved in multiple pathways within a protein interaction network may exert significant influence on the network and perform essential functional roles.

we aimed to prioritize nodes frequently visited by walkers along the numerous paths spanning from DIPs to DEPs.

DIP = Directly Interacting Proteins (with SARS-CoV-2, acute)
DEP = Differentially Expressed Proteins (also in acute illness)
with
PPI = Protein-Protein Interactions between these two layers

This approach is intended to uncover proteins central to COVID-19 that may significantly influence disease progression beyond the acute phase and contribute to the development of Long COVID.

So a lot of assumptions, including that the source data is a valid and accurate representation in the affected population.

To evaluate the potential of QW in identifying proteins from the SIP network, which go on to be associated with Long COVID, high-confidence Long COVID proteins (LCPs) were curated from […] Cervia-Hasler et al. 2024