Synthetically Engineered SARS-CoV-2 Spike Mutant Designed for Biocompatibility and Drug Resistance

In this study, I generated a synthetic SARS-CoV-2 Spike protein based on the XBB.1.5 (XFG) lineage, introducing 19 biologically plausible mutations aimed at modeling immune escape, drug resistance, and structural fitness. These mutations were selected to preserve viral function while exploring evolutionary pathways that could impact receptor binding, antibody recognition, and vaccine efficacy.

Using computational tools such as SWISS-MODEL and PyMOL, I validated the structural plausibility of the mutant and confirmed its similarity to natural variants in conformational space. Subsequently, I performed a Codon Pair Bias Index (CPBI) analysis across over 3,000 natural SARS-CoV-2 genomes and compared them with synthetic construct that I created for this study.

Surprisingly, the synthetic spike showed no significant deviation from the natural population in terms of codon pair frequency or adaptation index. Its CPBI score fell within the expected distribution of natural isolates, suggesting that engineered sequences designed under natural codon constraints may not be distinguishable from evolved ones using current genomic signature-based methods .

This work highlights the limitations of existing synthetic signal detection approaches , particularly when synthetic constructs are optimized for host codon compatibility and structural fidelity.

Biological Plausibility Over Engineering Signatures

The synthetic Spike variant was constructed using mutation strategies that:

Mimic known natural variation

Maintain physicochemical properties

Avoid rare codons or unnatural codon pairs

Preserve structural integrity and folding energy

As a result, despite being computationally engineered, the mutant exhibits codon pair frequencies consistent with natural isolates , making it indistinguishable from naturally occurring strains using standard CPBI metrics.

Implications for Biosecurity Detection

This suggests a critical limitation in current genomic surveillance techniques :

A synthetic virus designed with care to match natural codon usage can evade detection by CPBI-based algorithms.

This has important implications for:

Biosecurity monitoring

Synthetic origin detection

Future design of pathogen engineering safeguards

Conclusion: Rejection of CPBI as Sole Indicator

I conclude that CPBI alone cannot reliably distinguish engineered from naturally evolving sequences when synthetic designs are guided by biological constraints and codon optimization principles.

While CPBI remains a powerful tool for identifying obviously engineered constructs , its utility diminishes when synthetic genes are designed to mimic natural variation .

This opens new questions around:

How to define "unnatural" mutation. 

The need for multi-layered detection frameworks

As the sole author and researcher of this study and its accompanying data, I declare that I have no competing interests. This work has been designed to be fully reproducible, and the dataset and related materials are available on Zenodo. For additional files, collaboration opportunities, or to further develop this research, I can be contacted via tahirhb.com or by email at [e-mail]

References:Coleman et al., 2008

Coleman, J. R., Papamichail, D., Skiena, S., Futcher, B., Wimmer, E., & Mueller, S. (2008).Variable large-scale synthesis of genes: transgenes in plants as an example . Nucleic Acids Research , 36(2), e25.https://doi.org/10.1093/nar/gkn005 

Jackson et al., 2022

Jackson, C. B., Farzan, M., Chen, B., & Choe, H. (2022).Mechanisms of SARS-CoV-2 spike protein binding and entry into host cells . Annual Review of Medicine , 73, 39–54.https://doi.org/10.1146/annurev-med-041521-021418 

Waterhouse et al., 2018

Waterhouse, A., Bertoni, M., Bienert, S., Studer, G., Tauriello, G., Peer, G., ... & Schwede, T. (2018).SWISS-MODEL: homology modelling of protein structures and complexes . Nucleic Acids Research , 46(W1), W399–W404.https://doi.org/10.1093/nar/gky427 

Guex & Peitsch, 1997

Guex, N., & Peitsch, M. C. (1997).SWISS-MODEL and the Swiss-PdbViewer: An environment for comparative biomolecular modeling and visualization of molecular surfaces, dynamics and mutations . Electrophoresis , 18(15), 2714–2723.https://doi.org/10.1002/elps.1150181505 

Reference:Kelle et al., 2020

Kelle, A., Evans, N. G., & Altice, F. L. (2020).Dual-use risks of genome editing technologies: CRISPR and synthetic biology . Science and Public Policy , 47(5), 645–657.https://doi.org/10.1093/scipol/scz049 

Gronvall et al., 2021

Gronvall, G. K., & Kahn, J. S. (2021).Synthetic virology and global health security . Nature Reviews Microbiology , 19(4), 201–202.https://doi.org/10.1038/s41579-020-00483-z 

DeLano, 2002

DeLano, W. L. (2002).The PyMOL Molecular Graphics System . DeLano Scientific LLC , San Carlos, CA. 

Hadfield et al., 2018

Hadfield, J., Megill, C., Bell, S. M., Huddleston, J., Potter, B., Callender, C., ... & Bedford, T. (2018).Nextstrain: real-time tracking of pathogen evolution . Bioinformatics , 34(23), 4121–4123.https://doi.org/10.1093/bioinformatics/bty407

Shu & McCauley, 2017

Shu, Y., & McCauley, J. (2017).GISAID: Global initiative on sharing all influenza data – from vision to reality . Eurosurveillance , 22(13), 30494.https://doi.org/10.2807/1560-7917.ES.2017.22.13.30494

Clark et al., 2021

Clark, K., Coombes, B., Karsch-Mizrachi, I., Ostell, J., Sayers, E. W. (2021).GenBank . Nucleic Acids Research , 49(D1), D102–D106.https://doi.org/10.1093/nar/gkaa974 

Greaney et al., 2021

Greaney, A. J., Starr, T. N., Gilchuk, P., Zost, S. J., Binshtein, E., Loes, A. N., ... & Bloom, J. D. (2021).Comprehensive mapping of mutations in the SARS-CoV-2 receptor-binding domain that affect recognition by polyclonal human plasma antibodies . Nature Immunology , 22(12), 1493–1501.https://doi.org/10.1038/s41590-021-01038-9 

Starr et al., 2020

Starr, T. N., Greaney, A. J., Hilton, S. K., Ellis, D., Crawford, K. H. D., Navarro, M. J., ... & Bloom, J. D. (2020).Deep mutational scanning of SARS-CoV-2 receptor binding domain reveals constraints on folding and ACE2 binding . Cell , 182(5), 1294–1310.e20.https://doi.org/10.1016/j.cell.2020.08.050 

Shmakov et al., 2021

Shmakov, S. A., Yevshin, I. S., Sharipov, R. N., & Kolchanov, N. A. (2021).Designing synthetic DNA sequences with reduced bias in codon pair usage . BMC Bioinformatics , 22(1), 1–11.https://doi.org/10.1186/s12859-021-04482-9 

Jumper et al., 2021

Jumper, J., Evans, R., Pritzel, A., Green, T., Figurnov, M., Ronneberger, O., ... & Hassabis, D. (2021).Highly accurate protein structure prediction with AlphaFold2 . Nature , 596, 583–589.https://doi.org/10.1038/s41586-021-03819-2

Cock et al., 2009

Cock, P. J., Antao, T., Chang, J. T., Chapman, B. A., Cox, C. J., Dalke, A., ... & de Hoon, M. J. (2009).Biopython: freely available Python tools for computational molecular biology and bioinformatics . Bioinformatics , 25(11), 1422–1423.https://doi.org/10.1093/bioinformatics/btp163

Fecher & Friesike, 2014

Fecher, B., & Friesike, S. (2014).Open science: one term, five schools of thought . EPJ Data Science , 3(1), 1–12.https://doi.org/10.1140/epjds/s13111-014-0006-3