Critical Assessment of Massive Data Analysis (CAMDA) Conference Series

近年来的技术进步加速了生物医学研究的发展，为生物医学的发展提供了新的前景。 PACT在基础研究和应用或临床环境中的发现。计量数据采集卡的开发 被越来越流行的高通量分析所选择的，反过来又依赖于计算方面的创新- 国家生物学和生物信息学。至关重要的是，我们将需要进一步的方法学进步来检测 以及识别数据中具有生物医学相关性的新模式。这样的算法改进可以 考虑到缺乏真正新奇发现的基础事实，很难对其进行评估。然而，系统性的进步-- 呃依靠定期的绩效考核。为了满足研究界的这一公开需求，我们正在 举办了一系列关于海量数据分析关键评估(CAMDA)的会议，这些会议是 作为国际计算生物学学会(ISCB)ISMB年度会议的一部分， 该领域领先的专业组织。CAMDA已经成为一个著名的会议，专门讨论前 挖掘和推动生命科学中复杂数据分析的前沿。它最初成立于 2000(自然411,885.自然424,610)来提供对不同技术的批判性评估的论坛 用于大规模数据分析，包括但不限于高维基因表达谱，其他 -组学和临床数据(一般背景见www.camda.info)。它的目标是建立最先进的 在分析方法方面，确定进展和遗留问题，以突出有希望的方向 为未来的努力做准备。应对该领域的关键挑战，CAMDA是首批通过和 优化社区范围内竞赛的方法，科学界的竞争专家和 分解相同的数据集。CAMDA竞赛的重点是技术上正确的测量和信号处理 一方面，也是生物医学推理中要求最高的开放式问题。 CAMDA还与知名团体合作定义有效的比赛，挑战社区 对最新的最先进基准进行更深入的分析。例如，由 MAQC/SEQC联合体是突出的特征。这些财团由FDA的NCTR协调。在……里面 事实上，CAMDA竞赛包括与FDA的NCTR一起汇编的竞赛数据，从 从2012年开始，并定期进行。世界各地的研究人员被邀请接受CAMDA挑战，这是 已经成为一个重要的固定装置(参见自然方法5,569)，定期从 学术界和工业界。讨论了不同贡献分析的结果和方法，并进行了比较。 在会议上削减了开支。精选的演示文稿发表在《开放获取程序》特刊上， 与F1000 Research等领先的现代出版商合作，提供快速的公共传播 和Open Peer Review。代表们共同选出获胜的队伍和亚军。令人惊讶的是，最令人印象深刻的是- 积极的方法通常来自处于职业生涯早期的年轻科学家。

Recent technological advances have accelerated the development of biomedical research, promising high im- pact findings in both basic research and applied or clinical settings. Exploitation of the measurement data col- lected by the ever more prevalent high throughput assays, in turn, is dependent upon innovation in computa- tional biology and bioinformatics. Crucially, we will require further methodological advances for the detection and identification of novel patterns in the data that are of biomedical relevance. Such algorithmic advances can be hard to evaluate considering a lack of ground truth for truly novel discoveries. Systematic progress, howev- er, relies on regular performance evaluations. To address this open need of the research community, we are running a series of conferences on Critical Assessment of Massive Data Analysis (CAMDA) that are orga- nized as part of the annual ISMB meetings of the International Society for Computational Biology (ISCB), the leading professional organization in the field. CAMDA has become a renowned conference, specializing in ex- amining and driving the cutting edge of complex data analyses in the life sciences. It was originally founded in 2000 (Nature 411, 885. Nature 424, 610) to provide a forum for the critical assessment of different techniques used in large-scale data analysis including, but not limited to high-dimensional gene expression profiling, other -omics, and clinical data (see www.camda.info for general background). It aims to establish the state-of-the-art in analysis methods, identifying progress as well as remaining issues, in order to highlight promising directions for future efforts. Addressing key challenges in the field, CAMDA was one of the first conferences to adopt and optimize the approach of a community-wide contest, with competing experts of the scientific community ana- lysing the same data sets. CAMDA contests focus on technically correct measurements and signal processing on one hand, and the most demanding open ended questions of biomedical inference on the other hand. CAMDA also collaborates with high-profile groups on defining effective contests, challenging the community to deeper analyses of the latest state-of-the-art benchmarks. For instance, several benchmarks generated by the MAQC/SEQC consortia were featured prominently. These consortia were coordinated by the FDA’s NCTR. In fact, CAMDA competitions have included contest data compiled together with the FDA’s NCTR from the very beginning, and regularly since 2012. Researchers worldwide are invited to take the CAMDA challenge, which has already become a prominent fixture (cf. Nature Methods 5, 569), regularly drawing 60–100 specialists from academia and industry. The results and methods of the different contributed analyses are discussed and com- pared at the conference. Selected presentations are published in a special Open Access proceedings issue in collaboration with leading modern publishers, such as F1000 Research, offering fast-track public dissemination and Open Peer Review. Delegates jointly select a winning team and runner-ups. Strikingly, the most impres- sive approaches often come from young scientists in early stages of their careers.