Abstract
Molecular forces generated by cell receptors are infrequent and transient, and hence difficult to detect. Here we report an assay that leverages the CRISPR-associated protein 12a (Cas12a) to amplify the detection of cellular traction forces generated by as few as 50 adherent cells. The assay involves the immobilization of a DNA duplex modified with a ligand specific for a cell receptor. Traction forces of tens of piconewtons trigger the dehybridization of the duplex, exposing a cryptic Cas12-activating strand that sets off the indiscriminate Cas12-mediated cleavage of a fluorogenic reporter strand. We used the assay to perform hundreds of force measurements using human platelets from a single blood draw to extract individualized dose–response curves and half-maximal inhibitory concentrations for a panel of antiplatelet drugs. For seven patients who had undergone cardiopulmonary bypass, platelet dysfunction strongly correlated with the need for platelet transfusion to limit bleeding. The Cas12a-mediated detection of cellular traction forces may be used to assess cell state, and to screen for genes, cell-adhesion ligands, drugs or metabolites that modulate cell mechanics.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$29.99 / 30 days
cancel any time
Subscribe to this journal
Receive 12 digital issues and online access to articles
$99.00 per year
only $8.25 per issue
Buy this article
- Purchase on Springer Link
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
The data supporting the results in this study are available within the paper and its Supplementary Information. Raw data and images were deposited in Dataverse and can be accessed via the identifier https://doi.org/10.15139/S3/4Q8H1A. All raw and analysed datasets are available from the corresponding authors on request. No identifiable information from the participants will be disclosed. Source data are provided with this paper.
References
Orr, A. W., Helmke, B. P., Blackman, B. R. & Schwartz, M. A. Mechanisms of mechanotransduction. Dev. Cell 10, 11–20 (2006).
Zhang, Y. et al. Platelet integrins exhibit anisotropic mechanosensing and harness piconewton forces to mediate platelet aggregation. Proc. Natl Acad. Sci. USA 115, 325–330 (2018).
Ma, R. et al. DNA probes that store mechanical information reveal transient piconewton forces applied by T cells. Proc. Natl Acad. Sci. USA 116, 16949–16954 (2019).
Ramey-Ward, A. N., Su, H. & Salaita, K. Mechanical stimulation of adhesion receptors using light-responsive nanoparticle actuators enhances myogenesis. ACS Appl. Mater. Interfaces 12, 35903–35917 (2020).
Wang, X. & Ha, T. Defining single molecular forces required to activate integrin and notch signaling. Science 340, 991–994 (2013).
Gaudet, C. et al. Influence of type I collagen surface density on fibroblast spreading, motility, and contractility. Biophys. J. 85, 3329–3335 (2003).
Zhang, Y., Ge, C., Zhu, C. & Salaita, K. DNA-based digital tension probes reveal integrin forces during early cell adhesion. Nat. Commun. 5, 5167 (2014).
Rashid, S. A. et al. DNA tension probes show that cardiomyocyte maturation is sensitive to the piconewton traction forces transmitted by integrins. ACS Nano 16, 5335–5348 (2022).
Su, H. et al. Massively parallelized molecular force manipulation with on-demand thermal and optical control. JACS 143, 19466–19473 (2021).
Doudna, J. A. & Charpentier, E. The new frontier of genome engineering with CRISPR–Cas9. Science 346, 1258096 (2014).
Kaminski, M. M., Abudayyeh, O. O., Gootenberg, J. S., Zhang, F. & Collins, J. J. CRISPR-based diagnostics. Nat. Biomed. Eng. 5, 643–656 (2021).
Chen, J. S. et al. CRISPR–Cas12a target binding unleashes indiscriminate single-stranded DNase activity. Science 360, 436–439 (2018).
Gootenberg, J. S. et al. Multiplexed and portable nucleic acid detection platform with Cas13, Cas12a, and Csm6. Science 360, 439–444 (2018).
Broughton, J. P. et al. CRISPR–Cas12-based detection of SARS-CoV-2. Nat. Biotechnol. 38, 870–874 (2020).
Xiong, Y. et al. Functional DNA regulated CRISPR–Cas12a sensors for point-of-care diagnostics of non-nucleic-acid targets. J. Am. Chem. Soc. 142, 207–213 (2019).
Myers, D. R. et al. Single-platelet nanomechanics measured by high-throughput cytometry. Nat. Mater. 16, 230–235 (2017).
Ting, L. H. et al. Contractile forces in platelet aggregates under microfluidic shear gradients reflect platelet inhibition and bleeding risk. Nat. Commun. 10, 1204 (2019).
Cuisset, T. et al. Relationship between aspirin and clopidogrel responses in acute coronary syndrome and clinical predictors of non response. Thromb. Res. 123, 597–603 (2009).
Neira, H. D. & Herr, A. E. Kinetic analysis of enzymes immobilized in porous film arrays. Anal. Chem. 89, 10311–10320 (2017).
Duan, Y. et al. Mechanically triggered hybridization chain reaction. Angew. Chem. Int. Ed. 60, 19974–19981 (2021).
Glazier, R. et al. DNA mechanotechnology reveals that integrin receptors apply pN forces in podosomes on fluid substrates. Nat. Commun. 10, 4507 (2019).
Wang, X. et al. Integrin molecular tension within motile focal adhesions. Biophys. J. 109, 2259–2267 (2015).
Sahai, E., Ishizaki, T., Narumiya, S. & Treisman, R. Transformation mediated by RhoA requires activity of ROCK kinases. Curr. Biol. 9, 136–145 (1999).
Oba, M. et al. Cyclic RGD peptide-conjugated polyplex micelles as a targetable gene delivery system directed to cells possessing αvβ3 and αvβ5 integrins. Bioconjug. Chem. 18, 1415–1423 (2007).
Riikonen, T., Vihinen, P., Potila, M., Rettig, W. & Heino, J. Antibody against human α1β1 integrin inhibits HeLa cell adhesion to laminin and to type I, IV, and V collagens. Biochem. Biophys. Res. Commun. 209, 205–212 (1995).
Yamada, Y. et al. Structure-activity relationships of RGD-containing peptides in integrin αvβ5-mediated cell adhesion. ACS Omega 8, 4687–4693 (2023).
Belkin, A. M. et al. Transglutaminase-mediated oligomerization of the fibrin(ogen) αC domains promotes integrin-dependent cell adhesion and signaling. Blood 105, 3561–3568 (2005).
Roca-Cusachs, P., Gauthier, N. C., Del Rio, A. & Sheetz, M. P. Clustering of α5β1 integrins determines adhesion strength whereas αvβ3 and talin enable mechanotransduction. Proc. Natl Acad. Sci. USA 106, 16245–16250 (2009).
Galior, K., Liu, Y., Yehl, K., Vivek, S. & Salaita, K. Titin-based nanoparticle tension sensors map high-magnitude integrin forces within focal adhesions. Nano Lett. 16, 341–348 (2016).
Brockman, J. M. et al. Live-cell super-resolved PAINT imaging of piconewton cellular traction forces. Nat. Methods 17, 1018–1024 (2020).
Brockman, J. M. et al. Mapping the 3D orientation of piconewton integrin traction forces. Nat. Methods 15, 115–118 (2018).
Hu, Y. et al. DNA‐based microparticle tension sensors (μTS) for measuring cell mechanics in non‐planar geometries and for high‐throughput quantification. Angew. Chem. Int. Ed. 133, 18192–18198 (2021).
Dias, J. D. et al. Comparison of three common whole blood platelet function tests for in vitro P2Y12 induced platelet inhibition. J. Thromb. Thrombolysis 50, 135–143 (2020).
Le Blanc, J., Mullier, F., Vayne, C. & Lordkipanidzé, M. Advances in platelet function testing—light transmission aggregometry and beyond. J. Clin. Med. 9, 2636 (2020).
Davidson, B. L. et al. Bleeding risk of patients with acute venous thromboembolism taking nonsteroidal anti-inflammatory drugs or aspirin. JAMA Intern. Med. 174, 947–953 (2014).
Awtry, E. H. & Loscalzo, J. Aspirin. Circulation 101, 1206–1218 (2000).
Phillips, D. R. & Scarborough, R. M. Clinical pharmacology of eptifibatide. Am. J. Cardiol. 80, 11B–20B (1997).
Tcheng, J. E. et al. Pharmacodynamics of chimeric glycoprotein IIb/IIIa integrin antiplatelet antibody Fab 7E3 in high-risk coronary angioplasty. Circulation 90, 1757–1764 (1994).
Roth, G. & Majerus, P. W. The mechanism of the effect of aspirin on human platelets. I. Acetylation of a particulate fraction protein. J. Clin. 56, 624–632 (1975).
Tardiff, B. E. et al. Pharmacodynamics and pharmacokinetics of eptifibatide in patients with acute coronary syndromes. Circulation 104, 399–405 (2001).
Sassoli, P. M. et al. 7E3 F (ab′) 2, an effective antagonist of rat αIIbβ3 and αvβ3, blocks in vivo thrombus formation and in vitro angiogenesis. Thromb. Haemost. 85, 896–902 (2001).
Aungraheeta, R. et al. Inverse agonism at the P2Y12 receptor and ENT1 transporter blockade contribute to platelet inhibition by ticagrelor. Blood 128, 2717–2728 (2016).
Wang, X. et al. Comparative analysis of various platelet glycoprotein IIb/IIIa antagonists on shear-induced platelet activation and adhesion. J. Pharmacol. Exp. Ther. 303, 1114–1120 (2002).
Harder, S. et al. In vitro dose response to different GPIIb/IIIa-antagonists: inter-laboratory comparison of various platelet function tests. Thromb. Res. 102, 39–48 (2001).
De La Cruz, J., Bellido, I., Camara, S., Martos, F. & De La Cuesta, F. S. Effects of acetylsalicylic acid on platelet aggregation in male and female whole blood: an in vitro study. Scand. J. Haemotol. 36, 394–397 (1986).
Kirkby, N. et al. Antiplatelet effects of aspirin vary with level of P2Y (12) receptor blockade supplied by either ticagrelor or prasugrel. J. Thromb. Haemost. 9, 2103–2105 (2011).
Greiff, G. et al. Prediction of bleeding after cardiac surgery: comparison of model performances: a prospective observational study. J. Cardiothorac. Vasc. Anesth. 29, 311–319 (2015).
Dyke, C. et al. Universal definition of perioperative bleeding in adult cardiac surgery. J. Thorac. Cardiovasc. Surg. 147, 1458–1463 (2014).
Bartoszko, J. et al. Comparison of two major perioperative bleeding scores for cardiac surgery trials: universal definition of perioperative bleeding in cardiac surgery and european coronary artery bypass grafting bleeding severity grade. Anesthesiology 129, 1092–1100 (2018).
Kondo, C. et al. Platelet dysfunction during cardiopulmonary bypass surgery. With special reference to platelet membrane glycoproteins. ASAIO J. 39, M550–M553 (1993).
Roka-Moiia, Y. et al. Platelet dysfunction during mechanical circulatory support. Arterioscler. Thromb. Vasc. Biol. 41, 1319–1336 (2021).
Ranucci, M. et al. Platelet function after cardiac surgery and its association with severe postoperative bleeding: the PLATFORM study. Platelets 30, 908–914 (2019).
Welsh, K. J., Padilla, A., Dasgupta, A., Nguyen, A. N. D. & Wahed, A. Thromboelastography is a suboptimal test for determination of the underlying cause of bleeding associated with cardiopulmonary bypass and may not predict a hypercoagulable state. Am. J. Pathol. 142, 492–497 (2014).
Welsby, I. J. et al. The kaolin-activated thrombelastograph predicts bleeding after cardiac surgery. J. Cardiothorac. Vasc. Anesth. 20, 531–535 (2006).
Bowbrick, V. A., Mikhailidis, D. P. & Stansby, G. Influence of platelet count and activity on thromboelastography parameters. Platelets 14, 219–224 (2003).
Choi, J.-L., Li, S. & Han, J.-Y. Platelet function tests: a review of progresses in clinical application. BioMed. Res. Int. 2014, 456569 (2014).
Kaufmann, C. R., Dwyer, K. M., Crews, J. D., Dols, S. J. & Trask, A. L. Usefulness of thrombelastography in assessment of trauma patient coagulation. J. Trauma. 42, 716–720 (1997).
Moenen, F. C. et al. Screening for platelet function disorders with Multiplate and platelet function analyzer. Platelets 30, 81–87 (2019).
Quarterman, C., Shaw, M., Johnson, I. & Agarwal, S. Intra- and inter-centre standardisation of thromboelastography (TEG®). Anaesthesia 69, 883–890 (2014).
Harr, J. N. et al. Functional fibrinogen assay indicates that fibrinogen is critical in correcting abnormal clot strength following trauma. Shock 39, 45–49 (2013).
Slichter, S. J. et al. Dose of prophylactic platelet transfusions and prevention of hemorrhage. N. Engl. J. Med. 362, 600–613 (2010).
Femia, E. A., Scavone, M., Lecchi, A. & Cattaneo, M. Effect of platelet count on platelet aggregation measured by impedance aggregometry (Multiplate™ analyser) and by light transmission aggregometry. J. Thromb. Haemost. 11, 2193–2196 (2013).
Boknäs, N., Macwan, A. S., Södergren, A. L. & Ramström, S. Platelet function testing at low platelet counts: when can you trust your analysis? Res. Pract. Thromb. Haemost. 3, 285–290 (2019).
Greilich, P. E., Carr, M. E. Jr., Carr, S. L. & Chang, A. S. Reductions in platelet force development by cardiopulmonary bypass are associated with hemorrhage. Anesth. Analg. 80, 459–465 (1995).
Zwifelhofer, N. M. J. et al. Platelet function changes during neonatal cardiopulmonary bypass surgery: mechanistic basis and lack of correlation with excessive bleeding. Thromb. Haemost. 120, 94–106 (2020).
Holada, K., Šimák, J., Kučera, V., Rožňová, L. & Eckschlager, T. Platelet membrane receptors during short cardiopulmonary bypass—a flow cytometric study. Perfusion 11, 401–406 (1996).
Rinder, C. S. et al. Modulation of platelet surface adhesion receptors during cardiopulmonary bypass. Anesthesiology 75, 563–570 (1991).
Gouin, I. et al. In vitro effect of plasmin on human platelet function in plasma. Inhibition of aggregation caused by fibrinogenolysis. Circulation 85, 935–941 (1992).
Pareti, F. I., Capitanio, A., Mannucci, L., Ponticelli, C. & Mannucci, P. M. Acquired dysfunction due to the circulation of ‘exhausted’ platelets. Am. J. Med. 69, 235–240 (1980).
Guerrero, J. A. et al. Visualizing the von Willebrand factor/glycoprotein Ib–IX axis with a platelet-type von Willebrand disease mutation. Blood 114, 5541–5546 (2009).
Casari, C. et al. von Willebrand factor mutation promotes thrombocytopathy by inhibiting integrin αIIbβ3. J. Clin. Invest. 123, 5071–5081 (2013).
Andrews, R. K., López, J. & Berndt, M. C. Molecular mechanisms of platelet adhesion and activation. Int. J. Biochem. Cell Biol. 29, 91–105 (1997).
Chen, Y. et al. An integrin αIIbβ3 intermediate affinity state mediates biomechanical platelet aggregation. Nat. Mater. 18, 760–769 (2019).
Daniel, J. & Tuszynski, G. in Hemostasis and Thrombosis: Basic Principles and Clinical Practice 569–573 (JB Lippincott, 1994).
Oshinowo, O. T., Copeland, R., Bennett, C. M., Lam, W. A. & Myers, D. R. Platelet contraction force as a biophysical biomarker for bleeding risk in patients with immune thrombocytopenia. Blood 132, 517 (2018).
Wang, T. Y. et al. Cluster-randomized clinical trial examining the impact of platelet function testing on practice: the treatment with adenosine diphosphate receptor inhibitors: longitudinal assessment of treatment patterns and events after acute coronary syndrome prospective open label antiplatelet therapy study. Circ. Cardiovasc. Interv. 8, e001712 (2015).
Feghhi, S. et al. Glycoprotein Ib–IX–V complex transmits cytoskeletal forces that enhance platelet adhesion. Biophys. J. 111, 601–608 (2016).
Lee, J., Leonard, M., Oliver, T., Ishihara, A. & Jacobson, K. Traction forces generated by locomoting keratocytes. J. Cell Biol. 127, 1957–1964 (1994).
Liu, Y. et al. DNA-based nanoparticle tension sensors reveal that T-cell receptors transmit defined pN forces to their antigens for enhanced fidelity. Proc. Natl Acad. Sci. USA 113, 5610–5615 (2016).
Ma, R. et al. DNA tension probes to map the transient piconewton receptor forces by immune cells. J. Vis. Exp 20, e62348 (2021).
Gurbel, P. A. et al. First report of the point-of-care TEG: a technical validation study of the TEG-6S system. Platelets 27, 642–649 (2016).
Acknowledgements
We acknowledge support from NIH 5R01GM131099-04 (K.S.), NIH RM1GM145394 (K.S.) and NSF DMR 1905947 (K.S.). Y.D. is supported by an American Heart Association postdoctoral fellowship (23POST1028975). R.S. acknowledges support from Emory University Department of Anaesthesiology (internal funds). We thank the Emory Mass Spectrometry Center and F. Strobel and H. Ogasawara for ESI–mass spectrometry measurement to validate oligonucleotides. We also thank L. Downey and C. Maier for helpful discussions.
Author information
Authors and Affiliations
Contributions
Y.D. and K.S. conceived the project. Y.D. designed experiments, analysed data and compiled the figures. F.S. helped with TEG, aggregometry experiments and related discussions. Y.H., W.C. and R.L. helped with platelet purification and related discussion. Y.K. helped with the design of the experiments. F.S. and R.S. helped design the clinical studies and obtained related samples. Y.D. and K.S. wrote the manuscript. All authors contributed to revising the manuscript.
Corresponding authors
Ethics declarations
Competing interests
Y.D., Y.K., R.S. and K.S. are the inventors of a patent application (US Application No. 63/417,239). The remaining authors declare no competing interests.
Peer review
Peer review information
Nature Biomedical Engineering thanks Yunfeng Chen and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data
Extended Data Fig. 1 Comparison of surface density as function of linker length.
a) Scheme showing surface tethered activator sequences with different linker lengths. b) Plot of normalized surface density as a function of linker length. All biotinylated activators were incubated on streptavidin modified surfaces for 1hr at 100nM. Surface density was determined using fluorescence microscopy of fluorescently tagged activators and normalized to the activator density without linker. Error bar represents S.E.M from five independent surfaces.
Extended Data Fig. 2 Cas12a auto-cleavage of surface-tethered activator.
a) Schematic showing the MCATS process. Molecular traction forces mechanically melt the duplex probe and reveal an activator sequence that triggers Cas12a to cleave the linker of immobilized activators. b) Representative RICM, and duplex rupture (red) fluorescence images after platelets were incubated on concealed activator surface. Imaging was initiated after 1 hr after seeding platelets. The images compare the RICM cell spread area and duplex rupture signal before and after Cas12a/gRNA and reporter DNA were added for 1hr. Scale bar = 12 μm. c) Plot of mean fluorescence intensity beneath individual platelet before and after amplification. Center for the error bars is the mean fluorescence intensity under 10 individual platelets. Values are raw levels and were not background subtracted. Error bar represents S.D from 10 individual platelets. The p value is calculated by two-sided student t test. p<0.0001. d) Plot of spread area beneath individual platelets determined from RICM before and after amplification. Significance is calculated with two-sided student t test, p = 0.0002. Error bar represents S.D from 10 individual platelets.
Extended Data Fig. 3 Integrin subtype inhibition assay with MCATS.
a) Scheme showing the workflow of how the integrin inhibition assay was performed with MCATS. b-c) MCATS results of HeLa cell traction forces inhibited by different integrin monoclonal antibodies. Error bars represent standard error of the mean from n = 4 (Ttol = 12pN, panel b) and n = 5 (Ttol = 56pN, panel c) independent experiments. Significance is calculated by paired two-tailed t-test.
Extended Data Fig. 4 Measuring MCATS signal as a function of platelet number seeded on a surface.
a) Plot of MCATS signal as a function of seeded platelet number. Error bar represents SEM from n = 3 independent experiments b) Representative of RICM, duplex rupture (red) fluorescence images and bright field images of tension signal when seeding different number of cells on concealed activator surface. Scale bar = 10 μm.
Extended Data Fig. 5 Day-to-day variance of platelet tension test.
a) MCATS results of two healthy donors’ platelet tension signal from three individual blood draws performed on separate days. Center of error bar represents average of the measurements. Error bar represents S.D. from n = 2, 4, 2 for donor 1 on different days and 10, 4, 2 for donor 2 at different days.
Extended Data Fig. 6 LTA data for platelet treated with different ADP agonist concentrations.
a) Plot showing LTA signal of primary aggregation for platelets treated with ADP. LTA was accomplished using PAP-8E profiler prewarmed to 37 °C. The blank (0% aggregation) was set with PPP. b) Representative RICM, duplex rupture fluorescence (red) images for one donor washed platelets that were treated with ADP with concentrations ranging from 0.1 mM to 20 mM. Scale bar = 10 μm. MCATS signal for this data is shown in main Fig. 3.
Extended Data Fig. 7 MCATS measures dose-response curves for platelet inhibitors.
a–d) Plots of [Aspirin], [Eptifibatide], [7E3], [Ticagrelor] vs MCATS signal for different donors. A dose-response titration of six drug concentrations for each drug for individual donors is measured with MCATS and all measurements were performed in duplicate or triplicate. Mechano-IC50 for each donor was calculated by fitting plot to a standard dose-response function: Signal = Bottom + (Top-Bottom)/(1+([drug]/IC50). Solid line represents the fitting and dashed line represents 95% CI. Representative RICM, duplex rupture fluorescence (red) and zoom-in fluorescence images for one donor with drug concentrations ranging from 0 mM to 10 mM. Scale bar = 10 μm.
Extended Data Fig. 8 Sensitivity of aspirin and ticagrelor for patients before and after surgery.
a) Plots of [Aspirin] vs MCATS signal for patients before and after surgery. Error bars represent mean and SD from three replicate wells. Center of error bar represents average of the three measurements. Mechano-IC50 for each donor was calculated by fitting plot to a standard dose-response function: Signal = Bottom + (Top-Bottom)/(1+([drug]/IC50). The difference of calculated IC50 before and after surgery for two patients is non-significant with two tailed student t test, P = 0.65. b) Plots of [Ticagrelor] vs MCATS signal for patient before and after surgery. Mechano-IC50 was calculated by fitting plot to a standard dose-response function: Signal = Bottom + (Top-Bottom)/(1+([drug]/IC50).
Extended Data Fig. 9 Patient bleeding severity is correlated with platelet mechanical dysfunction.
Plot of reduction in MCATS signal (%) for subjects binned into two groups: mild/insignificant bleeding (class 1 and 2) or moderate/severe/massive bleeding (class 3, 4, and 5). Significance is calculated by a two tailed student t-test with P = 0.03. Error bars represent S.E.M. from n = 3 and 4 subjects respectively.
Extended Data Fig. 10 Tension measurement of lyophilized platelets.
a) Representative RICM, duplex rupture fluorescence (red) images for lyophilized platelets on tension probes. Scale bar represents 12 μm. b) Plot of MCATS signal of lyophilized platelets seeded on concealed DNA tension probes for 1hr. Error bar represents SD from 3 independent wells. Results indicate lyophilized platelets have no active mechanical signal.
Supplementary information
Source data
Source Data for Fig. 1
Source data.
Source Data for Fig. 2
Source data.
Source Data for Fig. 3
Source data.
Source Data for Fig. 4
Source data.
Source Data for Fig. 5
Source data.
Source Data for Extended Data Figs. 1–10
Source data for the extended data figures.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Duan, Y., Szlam, F., Hu, Y. et al. Detection of cellular traction forces via the force-triggered Cas12a-mediated catalytic cleavage of a fluorogenic reporter strand. Nat. Biomed. Eng 7, 1404–1418 (2023). https://doi.org/10.1038/s41551-023-01114-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41551-023-01114-1