Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Jun 26.
Published in final edited form as: Nat Genet. 2024 Nov;56(11):2287–2294. doi: 10.1038/s41588-024-01941-1

Pushing the boundaries of rare disease diagnostics with the help of the first Undiagnosed Hackathon

Angelica Maria Delgado-Vega 1,2, Helene Cederroth 3, Fulya Taylan 1,2, Katja Ekholm 1,2, Marlene Ek 1,2, Håkan Thonberg 1,2, Anders Jemt 4,5, Daniel Nilsson 1,6, Jesper Eisfeldt 1,6, Kristine Bilgrav Saether 6, Ida Höijer 7,8, Ozlem Akgun-Dogan 9,10,11, Yui Asano 12, Tahsin Stefan Barakat 13,14, Dominyka Batkovskyte 2, Gareth Baynam 15,16,17,18, Olaf Bodamer 19, Wanna Chetruengchai 20,21, Pádraic Corcoran 7,8, Madeline Couse 22, Daniel Danis 23, German Demidov 24, Eisuke Dohi 25, Mattias Erhardsson 26, Luis Fernandez-Luna 27, Toyofumi Fujiwara 28, Neha Garg 29, Roberto Giugliani 30,31,32,33, Claudia Gonzaga-Jauregui 27, Giedre Grigelioniene 1,2, Tudor Groza 15,34,35,36, Cecilia Gunnarsson 37, Anna Hammarsjö 1,2, Charles Kumi Hammond 38, Özden Hatirnaz Ng 10,11,39, Sirisha Hesketh 40, Dineshani Hettiarachchi 41, Maria Johansson Soller 1,2, Umn Ahmed Kirmani 42, Martin Kjellberg 43, Malin Kvarnung 1,2, Oleg Kvlividze 44,45, Kristina Lagerstedt-Robinson 1,2, Paul Lasko 46, Timo Lassmann 47, Lynette YS Lau 48, Steven Laurie 49,50, Weng Khong Lim 34,51,52,53, Zhandong Liu 54, Mariya Lysenkova Wiklander 8,55, Prince Makay 56, Helena Malmgren 1,2, Alassane Baneye Maiga 57, Carolina Maya-González 2, M Stephen Meyn 58,59, Ramprasad Neethiraj 4,60, Vincenzo Nigro 61,62, Felix Nordgren 2, Jessica Nordlund 8,55, Sara Orrsjö 63,64, Jesper Ottosson 63,64, Ugur Ozbek 10,65, Özkan Özdemir 10,39, Clyde Partin 66,67, David A Pearce 68,69,70,71, Raquel Peck 72, Annie Pedersen 63,64, Maria Pettersson 1,2, Monnat Pongpanich 73,74, Manuel Posada de la Paz 75,76, Arun Ramani 77, Juan Andres Romero 78, Vanessa I Romero 78, Richard Rosenquist 1,2,5, Aung Min Saw 79, Matthew Spencer 80,81, Eva-Lena Stattin 82, Chalurmpon Srichomthong 20,21, Isabel Tapia-Paez 83, Domenica Taruscio 84,85, Julie P Taylor 86, Tinatin Tkemaladze 87,88,89, Ian Tully 90, Zeynep Tümer 91,92, Wendy AG van Zelst-Stams 93, Alain Verloes 94,95, Emma Västerviga 96,97, Sailan Wang 83, Rachel Yang 98,99, Shinya Yamamoto 100,101, Vicente A Yépez 102, Qing Zhang 103, Vorasuk Shotelersuk 20,21, Samuel Agyei Wiafe 104, Yasemin Alanay 9,10,11, Lorenzo D Botto 105, Salman Kirmani 106, Aimé Lumaka 56,107,108, Elizabeth Emma Palmer 109,110, Ratna Dua Puri 29,111, Valtteri Wirta 1,4, Anna Lindstrand 1,2, Orion J Buske 112,#, Mikk Cederroth 3,#, Ann Nordgren 1,2,63,64,#
PMCID: PMC12198426  NIHMSID: NIHMS2083970  PMID: 39433890

Abstract

The first-ever Undiagnosed Hackathon was a groundbreaking event held by the Wilhelm Foundation, the Karolinska Undiagnosed Disease Program, and PhenoTips in collaboration with UDNI to solve medical mysteries and advance diagnostics for undiagnosed rare diseases. Nearly 100 healthcare professionals and researchers from 28 countries participated, working intensively for 48 hours to diagnose 10 families with undiagnosed rare diseases. This innovative approach to precision diagnostics highlighted the power of international, multidisciplinary collaboration and patient partnership, yielding promising results for patients seeking answers and benefiting the entire rare diseases community.

Keywords: rare disease diagnostics, undiagnosed diseases, genomic medicine, international collaboration, interdisciplinary research

Background

Undiagnosed rare diseases pose a significant burden to patients, families, healthcare professionals, and researchers worldwide1. Rare diseases collectively affect 3.5–5.9% of the global population2,3. Estimates suggest there are between 7,000 to 11,000 known rare diseases, with 72–80% believed to have a genetic basis2,3. Despite significant advances in genomic medicine, around 60% of patients with rare diseases remain undiagnosed even after comprehensive genetic testing4,5. Their diagnostic odysseys can span years, leading to significant emotional, financial and social burdens. They often face stigma, isolation, uncertainty, and delays in accessing therapies6. This situation is worsened by a lack of awareness and limited access to advanced diagnostic genomic technologies, especially in low- and middle-income countries (LMICs)7.

A timely and precise diagnosis can significantly improve patient care by providing a deeper understanding of potential symptoms or complications, guiding early decisions on follow-up and management options such as targeted therapies, risk (i.e. cancer) surveillance, genetic counselling, and prenatal diagnosis. The rapid advancement of precision therapies for previously untreatable rare diseases underscores the importance of accurate diagnoses8. Moreover, a diagnosis brings immense relief to patients and caregivers by providing knowledge about how the condition will affect their lives and enabling them to coordinate efforts within patient organizations.

Currently, hundreds of millions of people globally live with rare undiagnosed conditions, contributing to a significant public health challenge. Developing new diagnostic technologies and analysis tools that accurately detect all types of disease-causing variants, accelerating the rapid integration of these technologies into routine clinical practice, promoting global collaboration, and facilitating secure data sharing are critical steps to shorten their diagnostic odysseys9. A comprehensive catalogue of rare diseases, including phenotype and genotype data 10, supported by mandatory international cooperation frameworks, is required11. The Undiagnosed Diseases Network International (UDNI)12, aims to bridge the diagnostic gap by enhancing global diagnostic capabilities, especially in resource-limited settings, and fostering knowledge and technology exchange. Patient advocacy is central to the UDNI’s strategy, co-founded by the Wilhelm Foundation, a global organisation of patients with undiagnosed diseases, connecting the community, driving international collaboration and fostering innovation to find answers for patients and families.

The idea behind the First Undiagnosed Hackathon

A hackathon is an event where participants engage in intensive, collaborative problem-solving over a short period. Originally created as a tech innovation strategy, hackathons have expanded into other fields, including medicine and healthcare13,14, promoting interdisciplinary collaboration to solve complex challenges. An Undiagnosed Hackathon aligned with the global community’s vision to advance rare disease diagnosis and promote global collaboration while raising awareness for undiagnosed diseases.

In June 2023, the Wilhelm Foundation, the Karolinska Undiagnosed Diseases Program and PhenoTips, organized the first Undiagnosed Hackathon at Karolinska University Hospital and Karolinska Institutet (KI) in Sweden. Nearly 100 experts from 28 countries worked together for 48 hours on comprehensive genomic and phenotypic data from 10 undiagnosed families utilizing a myriad of computational tools and a multidisciplinary approach.

Participant selection

The UDNI’s Diagnostic Working Group (DWG) selected 13 undiagnosed patients from 10 families (three with two affected siblings) nominated by their local UDPs in China, the Democratic Republic of Congo, Ghana, India, Pakistan, Sweden, Turkey, and the United States of America. All of whom had previous diagnostic assessments through their local Undiagnosed Diseases Programs and remained undiagnosed. These patients were discussed on the UDNI-DWG forum and in virtual meetings. The organizers invited ~100 experts in rare diseases, including geneticists, clinicians, bioinformaticians, molecular biologists, scientists, stakeholders, and developers (Figure 1). Patients or their legal guardians provided informed consent.

Figure 1.

Figure 1.

Open in a new tab

Undiagnosed families from eight UDNI member countries were selected. After informed consent was obtained, blood samples were collected from patients and their families. DNA and RNA were sequenced using short-read whole genome and RNA sequencing, long-read genome and RNA sequencing and EPIC methylation array. Standardized deep phenotype data, together with the genomic data, were analyzed in multidisciplinary teams, which worked intensively for 48 hours to interpret the data. Four families received a diagnosis, candidate genes were found in four other families, and recommendations for further studies were provided for all undiagnosed cases. Undiagnosed Hackathon logo created by Yui Asano. Figure created with BioRender.com

Technologies

Blood samples were processed by the Clinical Genetics and Genomics Department at Karolinska University Hospital in Stockholm, Sweden. DNA and RNA were sequenced at the Clinical Genomics unit at SciLifeLab Stockholm and Uppsala nodes using advanced genomic technologies, including short-read whole genome sequencing (srWGS) (Illumina®), short-read RNA sequencing (srRNAseq) (Illumina®), long-read whole genome (lrWGS) and RNA sequencing (lrRNAseq) (Oxford Nanopore Technologies®) and methylation EPIC arrays (Illumina®) (Figure 1). Short-read whole exome sequencing (Illumina®), lrWGS and lrRNAseq (Pacific Biosciences®) data were available for the Chinese patient. Systematic and standardized phenotype data was collected using the open-source phenotyping tool tip2toe, developed by the Karolinska UDP in collaboration with the Wilhelm Foundation and the UDNI DWG (https://tip2toe.org).

IT Infrastructure

Participants accessed the extensive phenotype and genotype data via Karolinska Institute’s (KI) IaaS Virtual Server and an in-house open-source variant interpretation tool, SCOUT. Broadband speed and IT security were optimized for the event.

Interdisciplinary and intercultural teams

The Undiagnosed Hackathon was uniquely diverse and multidisciplinary, integrating the expertise of geneticists, clinicians, bioinformaticians, molecular biologists, patient organizations, developers, and stakeholders, including industry and the patients. Typically, these groups work in silos, but the Hackathon provided a unique opportunity for the intersection of these diverse disciplines to occur, leading to explorations of new ways and new angles to tackle the diagnostic odyssey. This was further enhanced by the participation of experts from LMICs who provided fruitful and important insights.

Participants were divided into five teams: Smileys, Dinosaurs, Animals, Space, and Unicorns. Each team, comprising approximately 20 participants from various disciplines and countries, was given 2–3 undiagnosed patients to solve with complete freedom to develop their own diagnostic solutions, including in-person clinical evaluation, analysis pipelines, variant prioritization and interpretation tools.

Social aspects

The event began with a welcome from the parents of the Swedish patients and their affected children and their unaffected siblings, instilling a profound sense of meaning from the start and reaffirming the “patient at the centre” ethos. The teams spent 48 hours analyzing data, consulting with each other, and directly engaging with the patients and their families. A psychologically safe space encouraged participants to express ideas, take risks and explore innovative pathways.

Outcomes of the event

The diagnostic success varied among teams, with four of ten families (40%) receiving a definitive diagnosis on the first day. Successful approaches and tools were rapidly shared among teams, and unsolved cases were redistributed for further analysis. These cases were significantly more challenging, making success at this stage more difficult. Participants worked tirelessly, some forgoing sleep to continue their analyses. Recommendations for further studies were generated for the undiagnosed, with a candidate gene identified in an additional four families.

The participants’ diverse backgrounds served as a remarkable platform for rich networking opportunities and knowledge transfer across disciplines and countries, fostering partnerships among diverse stakeholders and enhancing the overall value and impact of the hackathon.

The diagnoses made had a direct impact on patients’ management and genetic counselling. One patient was identified with Costello syndrome caused by a de novo variant in HRAS:c.35G>C (p.Gly12Ala), leading to the initiation of cardiac and cancer surveillance. Another patient, previously misdiagnosed with cerebral palsy due to prematurity and perinatal complications, was diagnosed with a neurodevelopmental disorder caused by a de novo variant in CUL3:c.2246T>C (p.Ile749Thr). A third patient was diagnosed with autosomal recessive TBCK syndrome due to compound heterozygosity for a structural variant TBCK:c.2060–6793_2235+427del, and a splicing variant TBCK:c.382–2A>G, where lrWGS and srRNAseq were crucial for the diagnosis. This diagnosis immediately made the patient suitable for starting treatment with BRAMINO®. Finally, a patient was identified as homozygous for a coding pathogenic variant in IGHMBP:c.1328G>A (p.Arg443His) causing spinal muscular atrophy with respiratory distress type 1 (SMARD1), which had been dismissed by commercial genetic testing with WES, emphasising the importance of considering the technical limitations of all tests.

Challenges

One challenge was the lack of integrated analysis solutions for complex multi-omics data, which limited the full exploitation of the data and the expertise present at the event. Efforts to diagnose the remaining unsolved families continued for the three months following the Hackathon, including monthly online conference meetings to discuss updates and next steps. Patients and parents found comfort in knowing that every possible effort was made to find a diagnosis.

Conclusions and Recommendations for Future Undiagnosed Hackathons

The Undiagnosed Hackathon demonstrated the potential of global collaboration, patient engagement, cutting-edge technology, and the multidisciplinary approach to addressing the diagnostic gap in rare diseases. It provided a platform for informal learning, problem-solving, and collaboration, fostering a sense of community within the UDNI and the rare diseases field.

Future Undiagnosed Hackathons can accelerate solutions in undiagnosed rare disease diagnostics, including developing clinical tools for multi-omics data analysis integration, streamlined variant assessment methods, data-sharing frameworks, advanced phenotyping tools integrated into Electronic Health Records, large-scale gene discovery programs, and AI and machine learning models for variant interpretation. The focused time frame and collaborative spirit of hackathons stimulate creativity and innovation, pushing the boundaries of genomic diagnostics beyond what is readily achievable in day-to-day healthcare. The resources generated can be used to accelerate rare disease research and foster collaborative scientific relationships.

The 48-hour time frame of the event might have limited the number of solved cases. However, the protected time solely focused on the diagnostic aim, a psychologically safe space, and the collaborative spirit helped brake down barriers between disciplines and improve workflow. It inspired people to work towards a specific aim and challenge themselves to give their best.

Furthermore, the event has opened doors for future initiatives aligning with the IRDiRC goal that undiagnosed individuals will enter a globally coordinated diagnostic and research pipeline 15. The lessons learned and the successes achieved indicate that this format holds substantial potential for achieving health equity for patients worldwide and for a future where undiagnosed diseases have become a thing of the past.

In summary, the Undiagnosed Hackathon has presented an inspiring model of multidisciplinary global and patient partnership, advancing the borders of diagnosis for undiagnosed rare diseases.

https://www.undiagnosedhackathon.org/

Acknowledgements

We express our deep gratitued to the patients and their families participating in the Hackathon. The Undiagnosed Hackathon was organized by the Wilhelm Foundation with grants from the Chan Zuckerberg Foundation and the Karolinska Undiagnosed Diseases Program with grants from the The Swedish Brain Foundation (FO2021–0177) and Hållsten Research Foundation. Sequencing flowcells were donated by Illumina ® and Oxford Nanopore Technologies®. The Undiagnosed Hackathon received scientific, technical, IT and administrative support from the Genomic Medicine Sweden initiative, the Department of Clinical Genetics and Genomics at Karolinska University Hospital, the Clinical Genomics and National Genomics Infrastructure Units in Stockholm and Uppsala funded by Science for Life Laboratory (SciLifeLab), the Knut and Alice Wallenberg Foundation and the Swedish Research Council, the SNIC/Uppsala Multidisciplinary Center for Advanced Computational Science and the UPPMAX computational infrastructure, and the IT departments at Region Stockholm and Karolinska Institutet. Special thanks to Astra Law, INQ Consulting, Blique Hotel, Euvic and ICA Brottbyhallen in Stockholm.

References

  • 1.Ferreira CR The burden of rare diseases. Am J Med Genet A 179, 885–892 (2019). [DOI] [PubMed] [Google Scholar]
  • 2.Smith CIE, Bergman P & Hagey DW Estimating the number of diseases - the concept of rare, ultra-rare, and hyper-rare. iScience 25, 104698 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Nguengang Wakap S et al. Estimating cumulative point prevalence of rare diseases: analysis of the Orphanet database. Eur J Hum Genet 28, 165–173 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Wright CF et al. Genomic Diagnosis of Rare Pediatric Disease in the United Kingdom and Ireland. N Engl J Med 388, 1559–1571 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Edsjo A et al. Building a precision medicine infrastructure at a national level: The Swedish experience. Camb Prism Precis Med 1, e15 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Baynam G, Gomez R & Jain R Stigma associated with genetic testing for rare diseases-causes and recommendations. Front Genet 15, 1335768 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Sciascia S et al. Unmet needs in countries participating in the undiagnosed diseases network international: an international survey considering national health care and economic indicators. Front Public Health 11, 1248260 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Hoytema van Konijnenburg EMM et al. Treatable inherited metabolic disorders causing intellectual disability: 2021 review and digital app. Orphanet J Rare Dis 16, 170 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Austin CP et al. Future of Rare Diseases Research 2017–2027: An IRDiRC Perspective. Clin Transl Sci 11, 21–27 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Rubinstein YR et al. The case for open science: rare diseases. JAMIA Open 3, 472–486 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Zurek B et al. Solve-RD: systematic pan-European data sharing and collaborative analysis to solve rare diseases. Eur J Hum Genet 29, 1325–1331 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Taruscio D et al. The Undiagnosed Diseases Network International: Five years and more! Mol Genet Metab 129, 243–254 (2020). [DOI] [PubMed] [Google Scholar]
  • 13.Undiagnosed-1: Collaborative, Patient Focused, Genomics Research Case. (2019). [Google Scholar]
  • 14.Hacking Health. (2019). [Google Scholar]
  • 15.Nabbout R et al. Towards the international interoperability of clinical research networks for rare diseases: recommendations from the IRDiRC Task Force. Orphanet J Rare Dis 18, 109 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES