An average structural proteome of the currently circulating monkeypox virus.
An early-outbreak response: 3,713 genome sequences sampled within a year of the 2022 multi-country MPX outbreak, distilled into a consensus structural proteome of 210 proteins, each modelled three ways and screened for mutations at the binding sites of tecovirimat and brincidofovir.
A new outbreak. No structural picture of what is circulating.
By April 2023, the 2022 MPX outbreak had spread to 110 countries, with 86,956 confirmed cases and 119 deaths. The molecular response needs a structural picture of what is actually in patients today, not what was sequenced in a reference genome decades ago. And it needs it fast enough to inform drug repurposing, vaccine choice and mutation monitoring.
Global, fast, novel
MPX moved from endemic regions of Africa to 110 countries within months. Each genome sequenced from a patient is slightly different. There is no time to wait for a single curated reference.
Reference, not consensus
The standard approach models a single reference genome. That misses the mutational reality of what is currently circulating, and gives a false sense of stability around drug-binding sites.
Will tecovirimat keep working?
Tecovirimat targets phospholipase F13; brincidofovir targets DNA polymerase. If either active site mutates under selection pressure, drug efficacy drops. That has to be tracked in 3D, not just in sequence.
Sequence consensus, folded three ways.
The study built a consensus genome from 3,713 high-quality MPX sequences sampled from patients within one year of the outbreak. Open reading frames were identified, filtered against the non-redundant protein database, reduced to 210 representative proteins, and each was modelled with three state-of-the-art structure-prediction methods. Mutational analysis was then mapped onto the structures, with drug-binding pockets as the priority readout.
Consensus genome
3,713 high-quality MPX genome sequences sampled within a year of the outbreak combined into a consensus, with average conservation per position retained.
ORF identification
10,580 candidate open reading frames extracted from the consensus and filtered against the non-redundant protein database, reducing to 1,079 putative ORFs and 210 representative proteins.
3D modelling
Each of the 210 representative proteins modelled three times, using complementary state-of-the-art structure-prediction methods, to give a per-protein structural ensemble.
Drug-site mutational analysis
Mutation events mapped onto the structures, with explicit focus on the binding pockets of tecovirimat (target F13 phospholipase) and brincidofovir (target DNA polymerase).
What the study delivered.
A real, sized response to an active outbreak: a structural proteome built from patient data, not a single reference, and shared openly so the public-health community could build on it.
Tecovirimat and brincidofovir, still on target?
The two small-molecule antivirals available against MPX target two viral proteins: phospholipase F13 and the viral DNA polymerase. The consensus-proteome view lets us locate the observed mutations in 3D and check whether they cluster near the drug-binding sites — a direct, structural early-warning signal for resistance.
Tecovirimat → phospholipase F13
F13 is the established tecovirimat target. The mutation map shows the residues where the circulating viruses differ from the reference, and whether they fall in or near the drug-binding pocket. A clean pocket is good news for continued tecovirimat use.
Brincidofovir → DNA polymerase
Brincidofovir targets the viral DNA polymerase. The same 3D mutation map is generated for the polymerase, with mutation frequencies labelled. If active-site residues drift, that is a structural argument for revisiting the dosing or the drug choice.
Continuing dynamic mutation monitoring within the structural proteome presented here is essential to timely predict possible physiological changes in the evolving virus.
From the paperThe same machinery that ran on SARS-CoV-2, retargeted in days.
The point of the consensus-proteome approach is speed: when a new outbreak appears, the pipeline that built the SARS-CoV-2 structural picture can be re-aimed at the next virus within days, not months.
Consensus genome. 3,713 high-quality patient-derived MPX sequences combined into a single consensus, with per-position conservation preserved as a quality signal.
ORF mining and filtering. 10,580 candidate ORFs reduced through non-redundant database filtering to 1,079, then to 210 representative proteins covering the reference structural proteome.
Triple-method 3D ensemble. Three state-of-the-art structure-prediction methods applied to each representative protein, producing per-protein structural ensembles instead of single brittle predictions.
Mutation overlay. The mutational landscape from the 3,713 genomes mapped back onto the 3D models, with frequencies labelled. Lets a public-health team see where mutations actually accumulate.
Drug-binding-site focus. Mutation overlay restricted to the binding sites of approved antivirals (tecovirimat F13; brincidofovir DNA polymerase) as a structural early-warning signal for emerging resistance.
Open data. Structural ensembles, ORF tables and mutation analyses released alongside the paper for direct reuse by other researchers and clinicians.
The paper.
AI-assisted structural consensus-proteome prediction of human monkeypox viruses isolated within a year after the 2022 multi-country outbreak
Parigger, Krassnigg, Grabuschnig, K. Gruber, Steinkellner & C. C. Gruber · Microbiology Spectrum 11(5), e02274-23 (2023) · American Society for Microbiology · doi: 10.1128/spectrum.02274-23
Innophore
Consensus genome assembly, ORF mining, triple-method 3D modelling pipeline, mutation overlay, drug-binding-site analysis. First and corresponding authorship.
Authors: Lena Parigger, Andreas Krassnigg, Stefan Grabuschnig, Georg Steinkellner, Christian C. Gruber.
University of Graz
Institute of Molecular Biosciences. Structural biology context and shared affiliations through the long-standing Innophore–Graz partnership.
Authors: Karl Gruber.
Bring us a new outbreak.
The consensus-proteome plus drug-target mutation pipeline runs on any virus where patient genomes are accumulating and approved antivirals exist. We can stand it up against your virus of interest within days.