Author List: Kyle McMahon1, Stella Nielsen1, Hannah Knoll1, Resham Talwar1, Davina Thompson1, Colby Wilkason1, Al Ozonoff1,2,3, Elyse Stachler1†, Pardis C. Sabeti1,4,5†
Author Affiliations:
1Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA
2Boston Children’s Hospital, Boston, Massachusetts, USA
3Harvard Medical School, Boston, Cambridge, Massachusetts, USA
4Harvard University, Cambridge, Cambridge, Massachusetts, USA
5Howard Hughes Medical Institute, Chevy Chase, Maryland, USA
†These authors have contributed equally to this work
Corresponding author: Elyse Stachler, estachle@broadinstitute.org
Summary
In May 2026, an outbreak of Bundibugyo ebolavirus (BDBV) in the Democratic Republic of the Congo was declared a Public Health Emergency of International Concern by the World Health Organization, underscoring the need for reliable, rapidly deployable diagnostic tools. Here, we present the development and analytical validation of reverse-transcriptase quantitative PCR (RT-qPCR) assays designed for sensitive and specific BDBV detection. We developed and evaluated three assay formats: a singleplex TaqMan assay, a duplex TaqMan assay incorporating a human internal control, and a singleplex SYBR assay to reduce dependence on probe availability during outbreak response. We evaluated assay performance using synthetic RNA gene fragments, whole viral RNA, and contrived clinical samples. Both TaqMan assays achieved a 95% limit of detection (LOD95) of 5 copies/reaction, while the SYBR Green assay achieved an LOD95 of 50 copies/reaction. In addition, the assays accurately detected BDBV viral RNA and did not detect Zaire ebolavirus or Sudan ebolavirus viral RNA under the conditions tested. By making the assay protocols and resources openly available through protocols.io, this assay set provides a flexible and accessible molecular detection tool for BDBV surveillance, research, and outbreak response.
Context
On May 5, 2026, the Ministry of Health of the Democratic Republic of the Congo (DRC) alerted the World Health Organization (WHO) to an outbreak of an unknown illness in Mongbwalu Health Zone, Ituri Province, following the deaths of four health workers within four days (WHO, 2026). On May 15, 2026, rapid response teams confirmed the outbreak as Ebolavirus Disease (EVD) caused by Bundibugyo virus (BDBV) (Democratic Republic of the Congo, Ministry of Public Health, 2026). Two days later the WHO declared the outbreak to be a Public Health Emergency of International Concern (PHEIC) due to the geographic spread in the DRC and Uganda, and a rapidly rising caseload (WHO, 2026). As of June 6, 2026, the DRC and Uganda Ministries of Health have reported 534 laboratory-confirmed cases and 93 confirmed deaths across the two countries (CDC, 2026). The rapid increase in suspected cases and fatalities has heightened public health concerns and underscored the need for reliable, rapidly deployable diagnostic tools.
Orthoebolavirus is a filovirus genus comprising six genetically distinct species of filamentous, linear, non-segmented negative sense, single-stranded RNA viruses (Cheng et al., 2025). Four species cause EVD in humans: Orthoebolavirus zairense, or Zaire ebolavirus (Z-EBOV); Orthoebolavirus sudanense, or Sudan ebolavirus (S-EBOV); Orthoebolavirus bundibugyoense, or Bundibugyo ebolavirus (BDBV); and Orthoebolavirus taiense, or Taï Forest ebolavirus (CI-EBOV) (CDC, 2026). The two remaining species circulate only in animal hosts: Orthoebolavirus restonense, or Reston ebolavirus (RESTV) and Orthoebolavirus bombaliense, or Bombali ebolavirus (BOMV) (CDC, 2026). The May 2026 outbreak is the 17th Ebola outbreak reported in the DRC since 1976 and the third outbreak globally attributed to BDBV. From 1976 through 2025, most reported EVD outbreaks, cases, and deaths were attributed to Z-EBOV, followed by S-EBOV, BDBV, and CI-EBOV (CDC, 2026).
Since Z-EBOV has caused the largest number of EVD outbreaks and has historically been associated with high case fatality rate, most commercially available, validated molecular diagnostics, vaccines, and therapeutics have focused on Z-EBOV, followed by S-EBOV and BDBV (Bettini et al., 2023 and WHO, 2024). This creates challenges during BDBV outbreaks, as assays designed for Z-EBOV and S-EBOV may not detect BDBV, allowing for continued early BDBV spread amongst communities (WHO, 2026). As patient cases continue to increase, significant gaps still remain for readily-available, robust diagnostics. Commercial reverse-transcription quantitative PCR (RT-qPCR) assays for BDBV detection are available, but open, non-proprietary options remain limited (Bettini et al., 2023). This creates challenges during a large and rapidly evolving outbreak, when proprietary designs may be difficult to independently assess against newly generated outbreak sequences and high demand may create supply chain constraints. Additionally, since these assays are proprietary, it is unknown if primer or probe mismatches to circulating viruses can also limit diagnostic performance and increase the risk of detection gaps as the outbreak evolves.
The limited availability of open, non-proprietary BDBV assay options motivated us to design RT-qPCR assays that could be optimized on both available sequences from the circulating outbreak strain and historical BDBV genomes, with the goal of creating a robust, transparent, and readily deployable detection tool. We implemented the assay in complementary formats suited to different operational settings: a high-sensitivity probe-based singleplex assay, a duplex assay that pairs viral detection with an internal human control, and a dye-based alternative that can be deployed when probe synthesis or supply is a limiting factor. We evaluated these formats using materials and workflows that can be rapidly reproduced by other laboratories, including synthetic RNA gene fragments, quantified standards, viral RNA, and contrived clinical samples generated in a human plasma background. We report the design process, analytical evaluation, and implementation protocols for these assays to support near-term outbreak response and future BDBV surveillance and research.
Results
In silico sensitivity and specificity analysis
We designed RT-qPCR primers and probes (sequences provided in Table 1) to detect BDBV and paired the viral assay with a human internal control assay adapted from Meddeb et al., 2019. The BDBV assay targets a 103 bp segment of the large polymerase (L) gene, which encodes the viral RNA-dependent RNA polymerase. At the time of design, the BDBV primers and probes have demonstrated 100% sequence identity to all 26 complete BDBV genomes available from NCBI Virus (Figure 1), as well as to 16 BDBV genome available through Pathoplexus as of June 4, 2026. In silico specificity analysis predicted no cross-reactivity of the BDBV primers and probes to Z-EBOV and S-EBOV genomes.
Table 1: Primer and probe sequences for Bundibugyo ebolavirus (BDBV) and mitochondrial circular DNA (mcirDNA) RT-qPCR assays. Target region, primer and probe names, sequences, and source for the BDBV assay and the human mcirDNA internal control assay.
Figure 1: Alignment of the Bundibugyo ebolavirus (BDBV) RT-qPCR primers and probe to historic BVBV genomes. Complete historical BDBV genomes were aligned in Geneious; a representative view of the alignment is shown for readability. The locations of the forward primer, probe, and reverse primer are indicated at the top of the alignment. Although sequence variation is present at some positions in the surrounding genomic region, all three primer and probe binding sites show 100% sequence identity across the historical genomes. The same primer and probe sequences also showed 100% sequence identity to available current outbreak genomes (n=16) as of June 4, 2026 (data not shown).
Analytical performance of TaqMan qPCR validated assays
We evaluated the TaqMan RT-qPCR assays using synthetic RNA gene fragments at known concentrations, whole viral RNA, and contrived clinical samples, defined here as whole viral RNA spiked into a normalized healthy human plasma background. The singleplex BDBV TaqMan assay exhibited an amplification efficiency of 96.6%, while the duplex BDBV assay exhibited an amplification efficiency of 93.9% (Figure 2). Both assays achieved a 95% limit of detection (LOD95) of 5 copies/reaction (Table 2). The human mitochondrial circular DNA (mcirDNA) internal control assay also performed similarly in singleplex and duplex formats, with amplification efficiencies of 103.5% and 104.1%, respectively. All standard curves had R2 values ≥ 0.99.
The BDBV assay also showed linear detection of whole viral BDBV RNA across serial dilutions in contrived clinical samples in both singleplex and duplex TaqMan formats (Figure 3, R2 ≥ 0.99). In addition, the BDBV TaqMan assay detected BDBV whole viral RNA and did not detect whole viral RNA from Z-EBOV or S-EBOV under the conditions tested. In the duplex assay, the mcirDNA internal control produced a consistent cycle threshold (Ct) across BDBV dilutions, as expected from the constant human plasma background. For all TaqMan assays, we selected 400nM forward primer, 400nM reverse primer, and 200nM probe as the optimal primer-probe concentration after down-selection based on standard curve efficiency within a target range of 90%–110%, linearity, and observed sensitivity using synthetic RNA gene fragment material.
Figure 2: Standard curves for Bundibugyo ebolavirus (BDBV) and mitochondrial circular DNA (mcirDNA) TaqMan RT-qPCR assays in singleplex and duplex formats. We generated standard curves for a) BDBV and b) the mcirDNA internal control assay using synthetic RNA gene fragments across serial dilutions. Data points show the mean and standard deviation of triplicate reactions. Solid lines show simple linear regressions, and dotted lines show the 95% confidence interval for the regressions. E indicates the RT-qPCR standard curve efficiency.
Table 1: Limit of detection (LOD) analysis for Bundibugyo ebolavirus (BDBV) RT-qPCR assays. We evaluated the 95% limit of detection (LOD95) for the singleplex TaqMan, duplex TaqMan, and SYBR Green BDBV assays using synthetic RNA gene fragments quantified by digital PCR. We defined the LOD95 as the lowest concentration that produced amplification in at least 95% of replicate reactions (n=21).
Figure 3: Detection of Bundibugyo ebolavirus (BDBV) whole viral RNA in contrived clinical samples. We generated contrived clinical samples by spiking BDBV whole viral RNA into normalized healthy human plasma background at three concentrations prepared by 10-fold serial dilutions of stock RNA. The BDBV assay showed linear detection across the dilution series in both singleplex and duplex formats. For the duplex assay, the mitochondrial circular DNA (mcirDNA) internal control showed a consistent cycle threshold (Ct) across BDBV dilutions, reflecting the constant human plasma background. Data points show the mean and standard deviation of triplicate reactions. Solid lines show simple linear regressions, and dotted lines show the 95% confidence intervals for the regressions.
Analytical performance of the SYBR Green RT-qPCR assay
We also evaluated the BDBV assay in a singleplex SYBR Green RT-qPCR format to provide an alternate detection modality that does not require a TaqMan probe, which can become a time-limiting reagent during rapid response scenarios. The SYBR Green assay had an amplification efficiency of 99.6% (R2 ≥ 0.99) and achieved an LOD95 of 50 copies/reaction (Figure 4, Table 2). We selected 150nM forward and reverse primer concentrations as the final condition after down-selection from tested concentrations based on standard curve efficiency within the target range of 90%–110%, linearity, and observed sensitivity using synthetic RNA gene fragment material. The SYBR Green assay also detected BDBV whole viral RNA in contrived clinical samples and did not detect whole viral RNA from Z-EBOV or S-EBOV, indicating the assay is still specific even without the added sequence discrimination provided by a TaqMan probe.
Figure 4: Standard curve for the Bundibugyo ebolavirus (BDBV) singleplex SYBR Green RT-qPCR assay. We generated the standard curve using synthetic RNA gene fragments across serial dilutions. Data points show the mean and standard deviation of triplicate reactions. The solid line shows a simple linear regression, and dotted lines show the 95% confidence interval for the regression. E indicates the RT-qPCR standard curve efficiency.
Conclusion
We present optimized and analytically validated RT-qPCR assays that expand open, non-proprietary options for sensitive and specific detection of BDBV. By implementing the assay as a singleplex TaqMan assay, a duplex TaqMan assay with a human plasma internal control, and a singleplex SYBR Green assay, we provide complementary formats that laboratories can adapt to different testing needs, reagent constraints, and outbreak-response settings. Across these formats, the assays achieved low limits of detection, detected BDBV RNA in contrived clinical samples, and did not detect Z-EBOV or S-EBOV RNA under the conditions tested. We provide protocols and assay resources through protocols.io to support rapid implementation by other laboratories. Together, these assays provide flexible tools for BDBV surveillance, outbreak response, and future research while reducing dependence on limited commercial assay options during high-demand public health emergencies.
Methods
Molecular assay design
Bundibugyo virus (BDBV) genomes (n=26) were downloaded from NCBI Virus with filters applied (Tax ID: 3052458, Nucleotide Completeness: complete, access date: May 17, 2026). These genomes were aligned in Geneious using MAFFT, and several TaqMan qPCR primer and probe combinations were designed using Geneious combined with Primer3. Once outbreak specific sequences were published, genomes were obtained from Pathoplexis (n=16 as of June 4, 2026). The chosen BDBV design had 100% sequence alignment to all historical BDBV sequences as well as to the newly released outbreak sequences. In addition, a previously published human internal control assay targeting circulating human mitochondrial DNA in human blood (mcirDNA) was adapted to a TaqMan qPCR assay and previously optimized (unpublished work). This assay was included as a human internal control for plasma samples.
In silico analysis of designs
Primers and probes were mapped to on-target and off-target sequences in Geneious Prime to test specificity in silico. An assay was predicted to detect a sequence if both primers and probe mapped with ≤ 3 mismatches. All assays were predicted to detect 100% of their on-target sequences and 0% of their off-target sequences. In addition, all primer and probe sequences were evaluated for broader cross-reactivity in NCBI BLAST.
Samples and controls
A synthetic double stranded DNA gene fragment (Twist Bioscience) representing the region of the BDBV genome the assay targets (L segment, RNA-dependent RNA polymerase) were in vitro transcribed (IVT) and DNase treated using the HiScribe® T7 High Yield RNA Synthesis Kit (New England Biolabs, E2040L) following manufacturer’s instructions. Post-IVT and DNase treatment, transcribed RNA material underwent purification using RNAClean XP beads (Beckman Colter, A63987) following manufacturer’s instructions. Purified RNA material was then quantified using Qubit™ RNA High Sensitivity (HS) Kit (Invitrogen, Q32852) as recommended by the manufacturer. Based on calculated quantification, the synthetic RNA material was diluted and normalized to 1E8 copies/𝜇L aliquots.
Synthetic double stranded DNA gene fragment (Twist Bioscience) targeting Mitochondrial Circular DNA (Cytochrome C Oxidase Subunit III) was used as a positive control for experimentation. This material was quantified using Qubit™ DNA High Sensitivity (HS) Kit (Invitrogen, Q33231) as recommended by the manufacturer. Based on calculated quantification, the synthetic DNA material was then diluted and normalized to 1E8 copies/𝜇L aliquots.
For simulated contrived sample experimentation whole viral RNA (Z-EBOV and S-EBOV provided by NEIDL, Boston University; BDBV obtained from BEI) was spiked into pooled extracted human plasma sample matrix obtained from healthy, non-infected individuals (Innovative Research). Briefly, human plasma was extracted utilizing the Quick-DNA/RNA MagBead Extraction Kit (Zymo Research, R2131) on the KingFisher™ Flex Magnetic Particle Processor (ThermoFisher) with the 96 Deep-Well Head as recommended by the manufacturer. This material was then pooled to create a standard and consistent human background signal in subsequent RT-qPCR experimentation.
Singleplex TaqMan qPCR primer-probe optimization
All assay designs underwent initial evaluation as singleplex FAM TaqMan qPCR assays at three different primer-probe concentrations (primer concentrations/probe concentration): 200nM/200nM, 400nM/200nM, and 600nM/200nM Probe (Integrated DNA Technologies). Evaluation and performance of these assays was conducted using the Luna Probe One-Step RT-qPCR Kit (No ROX) (NEB, E3007E) as 10𝜇L reactions in triplicate following manufacturer’s recommendations (see protocols for reaction composition and cycling conditions). Each primer-probe concentration was evaluated utilizing synthetic gene fragment RNA or DNA on a standard curve ranging from 1E7 copies/reaction–1E1 copies/reaction. Assay performance was then evaluated based on qPCR standard curve efficiency (ranging from 90%–110%), linearity, and observed sensitivity of the assay based on imputed material concentration. All qPCR experiments were conducted on a QuantStudio 6 Flex (Applied Biosystems).
Duplex optimization and evaluation
After selecting optimal primer-probe concentrations for each singleplex FAM TaqMan probe assay, a duplex assay composed of FAM (BDBV) and HEX (mcirDNA) was evaluated in singleplex and duplex form simultaneously utilizing each unique synthetic gene fragment RNA or DNA on a standard curve ranging from 1E7 copies/reaction–1E1 copies/reaction. Evaluation and performance of these assays was conducted using the Luna Probe One-Step RT-qPCR Kit (No ROX) (NEB, E3007E) as 10𝜇L reactions in triplicate following manufacturer’s recommendations. Duplex assay performance was compared to singleplex assay performance to ensure consistent sensitivity, qPCR standard curve efficiency, linearity, and general performance in a more complex, higher order multiplexing assay composition. All qPCR experiments were conducted on a QuantStudio 6 Flex (Applied Biosystems).
SYBR Green singleplex optimization and evaluation
A singleplex SYBR Green BDBV assay was optimized utilizing the following primer concentrations (forward and reverse primers at equimolar concentrations): 100nM, 150nM, 200nM, and 450nM (Integrated DNA Technologies). Evaluation and performance of this assay was conducted using the Power SYBR™ Green RNA-to-CT™ 1-Step Kit (Applied Biosystems™, 4389986) as 10𝜇L reactions in triplicate following manufacturer’s recommendations (see protocol for reaction composition and cycling conditions). Each condition was evaluated utilizing synthetic gene fragment RNA on a standard curve ranging from 1E7 copies/reaction to 1E1 copies/reaction. Assay performance was then evaluated based on qPCR standard curve efficiency (ranging from 90%–110%), linearity, and observed sensitivity of the assay based on imputed material concentration. All qPCR experiments were conducted on a QuantStudio 6 Flex (Applied Biosystems).
Limit of Detection Determination
To determine the limit of detection (LOD) of our optimized assays, we ran replicates (n=21) of the following synthetic gene fragment concentrations for the BDBV singleplex TaqMan assay, duplex assay, and SYBR assay: 5E1, 1E1, and 5E0 copies/reaction. The gene fragments were accurately quantified and normalized on digital PCR (dPCR) using the QIAcuity OneStep Advanced Probe Kit (Qiagen, 250131) following the manufacturer’s recommendations. The LOD95 was defined as the concentration where at least 20/21 replicates amplified. All qPCR experiments were conducted on a QuantStudio 6 Flex (Applied Biosystems).
Acknowledgements
The following reagent was obtained through BEI Resources, NIAID, NIH:
- RNA from Bundibugyo ebolavirus, Prototype Isolate #811250 (200706291 Uganda), NR-31812
We would like to thank the NEIDL at Boston University for providing RNA from Zaire ebolavirus and Sudan ebolavirus.
Funding Statement
This work is supported by the John D. and Catherine T. MacArthur Foundation, Flu Lab, and a cohort of generous donors through TED’s Audacious Project, including the ELMA Foundation, MacKenzie Scott, the Skoll Foundation, and Open Philanthropy.
Conflicts of Interest
P.C.S. holds several patents related to diagnostic technologies and is a co-founder and equity holder in Delve Biosciences and Lyra Labs, a board member and equity holder in Polaris Genomics, and an equity holder of NextGenJane. P.C.S was formerly a co-founder of Sherlock Biosciences and board member of Danaher Corporation, until December 2024. All potential conflicts are managed in accordance with institutional policy.
Supplemental Information
Table S1: Assay gene fragment sequences used as positive control sequences
Protocols (please note it may take a day or two for them to be public on protocol.io):
Appendix A: Protocol for Bundibugyo ebolavirus qPCR TaqMan singleplex assay
Appendix B: Protocol for Bundibugyo ebolavirus qPCR TaqMan duplex assay (Bundibugyo + human internal control)
Appendix C: Protocol for Bundibugyo ebolavirus qPCR SYBR singleplex assay