CARDIoGRAM celebrates its 10th Anniversary
Bibliographic record
Abstract
It is 10 years since CARDIoGRAM entered the scientific lexicon. It is the acronym for the Coronary ARtery DIsease Genome wide Replication and Meta-analysis, a global consortium that has set new standards in the exploration of the genetics of coronary artery disease (CAD) and myocardial infarction (MI).1,2 Even beyond these diseases, understanding the genetics of many other complex disorders and evaluation of the causal role of biomarkers associated with CAD have enormously benefited from the work of the consortium. The roots of CARDIoGRAM lie in an initial collaboration between the authors bringing together genome-wide association studies of CAD from the British Heart Foundation Family Heart Study (undertaken as part of the Wellcome Trust Case Control Consortium) and the German MI Family Heart Study.3 They recognized that much larger studies were needed to confidently identify genetic variants associated with CAD, given the modest effect size of individual variants and the statistical penalty created by the simultaneous analysis of half a million or more variants. Initially, 10 groups joined forces to create CARDIoGRAM. Founding principal investigators and respective studies of CARDIoGRAM were Nilesh J. Samani and Alistair Hall (WTCCC); Heribert Schunkert, Jeanette Erdmann, and Christian Hengstenberg (GERMIFS); Sekar Kathiresan (MIGen); Eric Boerwinkle, and Christopher J. O'Donnell, (CHARGE consortium); Robert Roberts, Ruth McPherson, and Alex Stewart (Ottawa Heart Study); Dan Rader and Muredach Reilly (PennCath/MedStar); Tom Quertermous and Tim Assimes (ADVANCE); Winfried März and Reijo Laaksonen (LURIC/AtheroRemo); Stefan Blankenberg (CADomics); and Unnur Thorsteinsdottir (deCODE). The first chairs were Samani and Schunkert (Figure 1). Heribert Schunkert, Jeanette Erdmann, Sekar Kathiresan, and Nilesh J Samani (from left to right) in 2008 at a Cardiogenics consortium meeting held in Lübeck, Germany. Three years thereafter, CARDIoGRAM merged with the C4D consortium4 and other groups in order to form CARDIoGRAMplusC4D, which since then has continued the discovery and functional exploration of genetic variants causing coronary heart disease.5,6 It was important for the founders to initiate a collaboration that is solely academic in nature. The only objective was to produce scientific information and detailed genomic data for exploration. It was the expressed intention to avoid transfer of any ownership between different parties joining the initiative in order to focus exclusively on the scientific work. At present work of the consortium has identified 164 chromosomal loci that contain genetic variants that are genome-wide significantly associated with CAD risk.7 Many more loci can be considered good candidates as their false discovery rate is below 5%.6 A foundation of this success was the principal decision to make aggregated data available to the scientific community. Thereby any researcher can download the summary statistics from the website of the consortium (www.cardiogramplusc4d.org) to study whether any specific genetic variant displays association with CAD. Using this valuable data set, a series of ever-larger meta-analyses detected multiple novel and rare genetic variants affecting the risk of CAD (Figure 2). Important publications and milestones achieved by CARDIoGRAM and CARDIoGRAMplusC4D. During these 10 years of joint research the number of loci with Genomewide significant association to coronary artery disease and myocardial infarction increased to 164 and likely will grow even further. To date, more than 70 publications have been published by the CARDIoGRAM consortium and its successor CARDIoGRAMplusC4D, many in leading journals such as, Nature, Nature Genetics, the New England Journal of Medicine, Lancet, and the European Heart Journal. But the unique resources created by CARDIoGRAMplusC4D were not only about the discovery of new loci causing CAD. The consortium set the stage for conducting Mendelian randomisation studies that by now have changed the perception of causal factors in the aetiology of CAD.8 The genetic risk for CAD mediated by most of the genetic variants discovered by CARDIoGRAM and its successor, CARDIoGRAMplusC4D, are repeatedly confirmed to be through unknown mechanisms emphasizing the opportunity for discovery of novel pathogenetic pathways other than the well-known cholesterol and other conventional networks.9,10 These observations have inspired researchers to pursue elucidation of these risk meditating pathways which will greatly contribute to the pathogenesis of CAD and provide a treasure trove of new targets for drug discovery and development.11 As an example, HDL-cholesterol is now questioned as a causal factor in the aetiology of CAD,12 while triglyceride-rich lipoproteins, or LP(a) are now established as being more than risk markers but rather factors that cause the disease.13–17 These findings are enormously important for informing the decision-making in drug development. Indeed, it is strongly recommended that directing therapeutic strategies to novel targets should be supported by genetic findings that document the causal roles of respective mechanisms.18 Likewise, the causal role of (classic) risk factors for atherosclerosis was explored in a new light as the genetic variants linked to hypertension, hypercholesterolaemia, smoking, telomere length, and many other traits were explored in the CARDIoGRAMplusC4D data set.19–22 Moreover, it became possible to explore systematically the overlap of CAD with other diseases such as large artery stroke,23 arterial aneurysms,24 or arterial dissection.25 Indeed, the genetic underpinnings of CAD were found in patients with heart failure, peripheral arterial disease, aortic stenosis, atrial fibrillation, and premature death to name the most relevant conditions in this respect.26,27 The same is true for anthropometric and social factors, as these may also play a role in the disease aetiology. For example, height28 or educational attainment29 were also seen in a different light after the respective genetic variants were studied for their association with CAD in the CARDIoGRAMplusC4D data set. The data accumulated by CARDIoGRAMplusC4D has also provided the basis for the construction of polygenic risk scores which are now creating a new paradigm for risk prediction of CAD.30 Such scores may have as much, if not greater clinical impact, than the development of new therapies based on the discoveries of the consortium. Beyond its scientific contributions the consortium may also be seen as a success story of the funding instruments of the European Union and other funders. Indeed, the EU-consortia Cardiogenics (coordinated by Heribert Schunkert, 2006–11) and Procardis (coordinated by Hugh Watkins, 2007–11), and AtheroRemo (coordinated by Reijo Laaksonen, 2008–13) formed the core of what became CARDIoGRAMplusC4D in 2011. Even today, the groups meet on bi-monthly calls (chaired by Erdmann and Samani) and at international meetings whenever possible. The photo shows Samani, Schunkert, Erdmann, and Kathiresan at a meeting in 2008 in Lübeck. Finally, beyond the pure science, CARDIoGRAM and CARDIoGRAMplusC4D are examples for the power of multinational collaborations as opposed to attempts by individual groups to solve a task. Conflict of interest: none declared. References are available as supplementary material at European Heart Journal online.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.011 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.008 | 0.003 |
| Open science | 0.002 | 0.005 |
| Research integrity | 0.005 | 0.007 |
| Insufficient payload (model declined to judge) | 0.133 | 0.128 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".