At LinkedIn, WhereHows currently stores data about the status of 50,000 datasets, 14,000 comments and 35 million job executions.
The outcome of interest, i.e, the case control status for Dataset 2, and tumor and normal status for Dataset 3, was incorporated into the batch correction model so that the biological information was taken into account.
However, as data volumes increase over the lifetime of a project, it can be increasingly hard to track the origin, status and location of datasets in such systems.
The principal aim of this study was to reassess histopathological predictors of BRCA1 and BRCA2 mutation status by analysis of datasets considerably larger than those analyzed previously for this purpose, to provide more robust pathology-based likelihood ratios for use in assessing the pathogenicity of BRCA1 or BRCA2 variants.
Gene set features were selected from one dataset and used to predict metastasis status of samples from a second validation dataset.
A Monte Carlo cross-validation simulation was set up to create training sets from 50% of the data and use the sets to predict tumor/normal status of the remainder of the dataset with 1000 iterations.
Subsequently, the condition was used to predict the remaining samples by training set A. To predict the status of the phenotype on a dataset, a Bayesian probit regression model was fit to assign the probability that the sample exhibited evidence of a phenotype, based on the concordance of its gene expression values with the signature (Gatza et al, 2010).
Although some correlations may be specific to cancer status, this dataset represents a large collection of different tissue types.
The priority date status of first line drugs in the dataset is reported in Table 4.
Due to the size of dataset 1, only one, although representative, distribution per immunological status is shown.
