Extensive SARS-CoV-2 spike sequence diversification converges on a bounded, low-dimensional physicochemical space

During the COVID-19 pandemic, SARS-CoV-2 accumulated extensive amino acid variation in its spike protein. However, the extent to which sequence diversification translates into physicochemical diversity remains unclear. Here, we analyzed 271,769 unique spike protein sequences from 12 geographical regions using three physicochemical descriptors: net charge at pH 7.0, hydropathy, and aliphatic index. To characterize temporal changes in physicochemical space, we compared the squared Mahalanobis distances of observed sequences with those of Monte Carlo-simulated sequences containing randomized amino acid substitutions, constrained to the number of empirically variable positions. We applied trimmed clustering to characterize sequence distributions in physicochemical space and used Shannon entropy to assess spatiotemporal relationships between genotypic and physicochemical diversity across geographic regions. Empirical physicochemical displacement increased early in the observation period and subsequently stabilized, despite continued increases in variable amino acid positions and accumulated substitutions. In contrast, simulated trajectories exhibited greater and sustained physicochemical divergence. Trimmed clustering identified six distinct physicochemical groups repeatedly occupied by genetically divergent lineages differing in mutation burden, amino acid composition, and emergence time. Shannon entropy analyses revealed broadly similar nonlinear temporal patterns across geographic regions, with differences between genotypic and physicochemical entropy tending to be negative. These findings indicate that extensive spike protein sequence diversification occurred within a comparatively restricted physicochemical space, consistent with a many-to-one relationship between amino acid sequences and physicochemical states. This framework provides a quantitative approach for investigating physicochemical constraints on viral protein evolution and may inform future machine learning approaches to epitope prediction and vaccine antigen design.

Authors

Institutions

Publication Details

Journal
bioRxiv (Cold Spring Harbor Laboratory)
Published
2026-11-16
DOI
https://doi.org/10.64898/2026.11.12.766201
Primary Topic
SARS-CoV-2 and COVID-19 Research
Type
preprint

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Extensive SARS-CoV-2 spike sequence diversification converges on a bounded, low-dimensional physicochemical space

Marianna Manoussopoulou, Cleo G. Anastassopoulou, Αθανάσιος Τσακρής, Julian A. Hiscox et al.
bioRxiv (Cold Spring Harbor Laboratory)
SARS-CoV-2 and COVID-19 Research
preprint

Extensive SARS-CoV-2 spike sequence diversification converges on a bounded, low-dimensional physicochemical space

Marianna Manoussopoulou, Cleo G. Anastassopoulou, Αθανάσιος Τσακρής, Julian A. Hiscox, John Ioannidis, Constantinos Siettos, Yiannis Manoussopoulos
preprint en

Abstract

During the COVID-19 pandemic, SARS-CoV-2 accumulated extensive amino acid variation in its spike protein. However, the extent to which sequence diversification translates into physicochemical diversity remains unclear. Here, we analyzed 271,769 unique spike protein sequences from 12 geographical regions using three physicochemical descriptors: net charge at pH 7.0, hydropathy, and aliphatic index. To characterize temporal changes in physicochemical space, we compared the squared Mahalanobis distances of observed sequences with those of Monte Carlo-simulated sequences containing randomized amino acid substitutions, constrained to the number of empirically variable positions. We applied trimmed clustering to characterize sequence distributions in physicochemical space and used Shannon entropy to assess spatiotemporal relationships between genotypic and physicochemical diversity across geographic regions. Empirical physicochemical displacement increased early in the observation period and subsequently stabilized, despite continued increases in variable amino acid positions and accumulated substitutions. In contrast, simulated trajectories exhibited greater and sustained physicochemical divergence. Trimmed clustering identified six distinct physicochemical groups repeatedly occupied by genetically divergent lineages differing in mutation burden, amino acid composition, and emergence time. Shannon entropy analyses revealed broadly similar nonlinear temporal patterns across geographic regions, with differences between genotypic and physicochemical entropy tending to be negative. These findings indicate that extensive spike protein sequence diversification occurred within a comparatively restricted physicochemical space, consistent with a many-to-one relationship between amino acid sequences and physicochemical states. This framework provides a quantitative approach for investigating physicochemical constraints on viral protein evolution and may inform future machine learning approaches to epitope prediction and vaccine antigen design.

bioRxiv (Cold Spring Harbor Laboratory)
University of Liverpool (GB), National and Kapodistrian University of Athens (GR), University of Naples Federico II (IT), Stanford University (US)
Medical Research Council
SARS-CoV-2 and COVID-19 Research
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon