SADI, SHARE, and the in silico scientific method.
SADI, SHARE, and the in silico scientific method.
复制标题
DOI：
10.1186/1471-2105-11-s12-s7
复制
发表时间：
2010-12-21
影响因子：
3
通讯作者：
Samadian S
中科院分区：
文献类型：
--
作者：
Wilkinson MD;McCarthy L;Vandervalk B;Withers D;Kawas E;Samadian S
语义网技术的出现和生命科学的吸收为探索计算机科学的新方法提供了令人兴奋的机会。Web服务工作流已经成为“新方式”中的第一类对象，并作为实验如何完成的显式，可共享，可引用的表示。反过来，语义Web服务项目的目的是促进工作流建设的生物领域的专家，工作流可以编辑，重新定位，并重新发布的非信息学家。然而，科学方法中与明确的论述、分歧和假设生成有关的方面仍然相对不受新技术的影响。在这里，我们提出了SADI和SHARE -一种新的语义Web服务框架，以及其客户端库的参考实现。SADI和SHARE一起允许半自动或全自动地发现和流水线化语义Web服务，以响应特定的用户查询。SADI和SHARE所展示的语义行为扩展了描述逻辑推理器所提供的功能，使得新颖的断言可以自动添加到数据集中，而无需逻辑推理，而是通过分析或注释服务。这种行为可能会被应用到实现“语义化”的那些方面，在silico的科学方法，还没有支持语义Web技术。我们使用临床研究领域的一个例子来支持这一建议。
The emergence and uptake of Semantic Web technologies by the Life Sciences provides exciting opportunities for exploring novel ways to conduct in silico science. Web Service Workflows are already becoming first-class objects in “the new way”, and serve as explicit, shareable, referenceable representations of how an experiment was done. In turn, Semantic Web Service projects aim to facilitate workflow construction by biological domain-experts such that workflows can be edited, re-purposed, and re-published by non-informaticians. However the aspects of the scientific method relating to explicit discourse, disagreement, and hypothesis generation have remained relatively impervious to new technologies. Here we present SADI and SHARE - a novel Semantic Web Service framework, and a reference implementation of its client libraries. Together, SADI and SHARE allow the semi- or fully-automatic discovery and pipelining of Semantic Web Services in response to ad hoc user queries. The semantic behaviours exhibited by SADI and SHARE extend the functionalities provided by Description Logic Reasoners such that novel assertions can be automatically added to a data-set without logical reasoning, but rather by analytical or annotative services. This behaviour might be applied to achieve the “semantification” of those aspects of the in silico scientific method that are not yet supported by Semantic Web technologies. We support this suggestion using an example in the clinical research space.