Microarray Game

There are many possibilities to analyze molecular Signatures. They can be compared to each other, meta-signature via Meta-Analysis can be derived, functional enriched terms can be assessed as well as Lifespan Factors identified.

It is possible to make such analyses just like a game. In its essence a molecular Signature can be treated basically as a list of Factors. So even non-scientist can play with them and see what they can discover. However, it needs to be very visually and interactive as well as intuitive.

In order to make this possible first of all identify all aging-related micro-array Datasets References. Classify them as Signatures

Include online resources (e.g. GEO, ArrayExpress) on gene expression data in the Links app to establish connection to this platforms. Classify them as online resource and Signatures.

It may also possible to upload specific Profiles and Signatures on to Denigma in order to transform them into a unified format amenable to computation and comparisons. Though the procedure and best format need to be established to make this scalable.

The first step is that we can make finding and classifying all Aging-relevant gene Expressions Datasets possible and straightforward. This should already be like a game as we seek to provide an up-to-date assembly of all Aging gene expression Signatures for completeness.

Identifying and correctly classifying a dataset will be merited with an Achievement system and therefore Encouraging Initiative. This is the first step towards a game. It shall provide the feeling of "Catch 'em all!".

Then we can see how we can obtain Datasets automatically and transform all of those into a common unified format.

Finally we will enable to analyze signatures via different means in an interactive and a visual attractive fashion. It shall be possible to perform virtual experiments. Users should be able to pick some sets of data where the expression of Lifespan Factors or Factors of interest are significant different (e.g. 2-fold or even 5-fold). By comparing those Datasets it will be possible to see the effect of overexpression/underexpression of such Factors on expression activity patterns. There need to be views that provide the possibility to cluster results and generate heatmaps.

The Expressions Application will provide a search form view that lets one query gene symbols/names/identifiers. It will return a list of signatures in which the Factors are significant over- or under-expressed beyond a defined cutoff. Next one can select those that appear to make sense and/or are interesting and apply clustering of Factors in those signatures which will result in the generation of an interactive heatmap.

From this students/teenagers can try to formulate hypotheses and get the possibility to test them even at home via Distributed Science. o


Tags: expression, signatures, analysis, profiles
Update | Engage

Comment on This Data Unit