What is Skim?

firefox

Skim stands for 'Simple knowledgebase for insightful metabolomics'. Skim is a is a free, open-source framework for simplifying interpretation of high throughput metabolomics data. There is a considerable gap between our ability to produce metabolomics data and our ability to analyze it to gain systems-level insight into the biological processes that produced the data. By pooling our collective insights into metabolomics and by developing novel tools to query, aggregate and analyze this collective pool of knowledge, we can address the sore lack of tools for producing biologically relevant knowledge based on quantitative information.

Skim is currently under testing. Feel free to play around, but features, layouts and other things may be altered considerably. Feel free to leave a notebook post with the tag "feature request" to suggest new features!

Technologies underlying Skim

Skim relies on two sources of metabolite, reaction and pathway information: Highly curated databases such as the KEGG PATHWAY Database and valuable insights and inputs provided by users such as yourself.

In addition to serving as a repository for your thoughts, opinions and conjectures, we are also developing tools by which you could query yours (and others') inputs. We envision these tools will go well beyond simple word-based searches. They will incorporate advanced natural language parsers to determine the structure of user inputs and provide a rapid overview of the knowledge base. The Skim framework will eventually provide an API that would allow you to develop your own methods for querying and analyzing the information in Skim.

 

Document-based information

Skim does not use a traditional (relational) database. It is built on top of CouchDB, a document-based database. This database provides unparalleled scalability and flexibility in the development of Skim. The kinds of tools and data sources relevant to metabolomics are continuously evolving. We do not know how your data analysis and storage needs evolve. CouchDB was built with exactly such a scenario in mind.

Every piece of information entered or queried as part of the Skim framework, (including this HTML page) is stored as a document. It is accessible over pure HTTP in a RESTful manner. These documents can be schema-free, and range from fully structured (quantitative data tables) to fully unstructured (quick notes in your notebook).

Because of the REST interface and the document-based model, replication of the database (and all its applications) is trivial. You can install your own CouchDB server and replicate this entire website locally if you wish.

Javascript-based tools

Applications built on CouchDB are built using pure Javascript and nothing else. We hope this considerably reduces the barrier potential developers to develop and share new analytical tools with the metabolomics community.

That said, the knowledge of Javascript is not essential to using the Skim knowledgbase. There are plenty of tools we and others will develop over time for you to focus on the task that you enjoy most: making sense of metabolomics.

Featured Insights

Lorem ipsum dolor sit

firefox Vestibulum venenatis. Nulla vel ipsum. Proin rutrum, urna sit amet bibendum pellentesque, ante sapien varius turpis, eu nonummy nunc urna scelerisque ante. Nunc sed leo at dolor pharetra facilisis. Sed sed lectus at justo dapibus tristique. Morbi molestie pulvinar elit. Donec dictum nibh nec est. Aliquam sagittis, lacus a pharetra tempor, enim justo tempor nisi, at aliquet nulla metus sed augue.

Latest tool

Notebook

Currently just a blog implementation. Comes with tagging and word clouds. This notebook will feature a mechanism by you could follow other authors (which you specify) in your feed stream

kbhalerao

Latest insight

New insights on glucose

Vestibulum venenatis. Nulla vel ipsum. Proin rutrum, urna sit amet bibendum pellentesque, ante sapien varius turpis, eu nonummy nunc urna scelerisque ante. Nunc sed leo at dolor pharetra facilisis. Sed sed lectus at justo dapibus tristique. Morbi molestie pulvinar elit. Donec dictum nibh nec est. Aliquam sagittis, lacus a pharetra tempor, enim justo tempor nisi, at aliquet nulla metus sed augue.