We chose to study pediatric ALL because there is already a substantial pool of datasets freely available in the public domain.
The SCORE system was developed by the SCORE project group [ 8] who assembled a pool of datasets from 12 European cohort studies in order to calculate separate risk functions for high and low risk regions of Europe.
The generation of a transcriptome map of the relative 'A'/'B' ratio expression, allows the easy investigation of regional differential expression without the need to generate the results separately for the two datasets or pool of datasets and to devise additional calculations to compare them.
Based on a pool of datasets from 12 European cohort studies, mainly carried out in general population settings, European experts performed the Systematic COronary Risk Evaluation (SCORE) project to develop a risk scoring system for use in the clinical management of CV risk in European clinical practice [ 12].
Thus, on the artificial data there is a clear synergetic effect associated with pooling of datasets.
After imputation and pooling of datasets, the age- and gender 10-year risk estimates for CHD increased for both groups (South Asians: 20.9 (19.7-22.1)%, Norwegians 19.9 (18.7-21.1)%, but ethnic differences remained insignificant.
Finally, a powerful feature of TRAM is its ability to compare, within the same analysis, the transcriptome maps derived from two datasets (or two pools of datasets) related to different biological conditions (indicated as 'A' and 'B'), such as two tissues or cell types, two developmental stages, normal vs pathological cells or cells maintained in absence or in presence of a substance.
Scaling factor values between 1.0 × 10-5 to 9.0 × 10-6 generated a pool of 500 datasets, and from these we arbitrarily selected 35 datasets with S-values of 5, 10, 15, 20, 25, 30 or 35 (five datasets each; Table 2; Additional file 2).
Starting with our pool of nine datasets (SLO, ISI, AMG, NOV, KAT, SHA, SAE, SIR, PHI), all datasets were first checked for uniqueness and then cross-checked against each other to retrieve a set of unique and independent datasets.
In general, however, when coupled with one of several cross-experiment dataset integration tools, aRrayLasso will enable mining of the remarkable and untapped historical pool of microarray datasets for large-scale metastudies for well-powered discovery.
As before the enrichment scores were calculated using the hypergeometric distribution and the values were normalized using the background pool of transcriptome datasets.
