To inform our predictions we utilize locality, the past demand trends of venues in our neighborhood, and temporal similarity, the demand trends of temporally synchronous areas of the city.
To prove this prediction, we utilized an ERP technique that measured both stimuli (sentence) and response (action) cortical processes.
Because our data sets did not include sufficient numbers of test compounds for derivation of training and validation nor did they include test data sets to appropriately perform a prediction, we utilized a phenotypic anchoring approach.
To develop prediction models, we utilized likelihood cross-validated penalized logistic regression models which implemented either an L1 penalty (Lasso) [22] or an L2 penalty (Ridge) using the R package 'penalized' [23].
To start addressing the validity of these predictions we first utilized previously reported experimental data for mouse LS cells [ 17], hybridoma cells [ 18], and a mixture of cancer and normal cells [ 4], indicating that cancer cells and normal cells follow the same law.
To minimize the number of erroneous protein predictions, we only utilized sequences if they fell into one of two categories: 1) significant similarity to a known protein sequence (BLASTX to NCBI's nr, evalue ≤1×10-5 ≤1×10-5predicted prorein sequence ≥200 amino acids long.
To further test the predictions by our model, we utilized fluorescence resonance energy transfer-based lipid fusion assays, in which neuronal SNAREs are reconstituted into liposomes at physiologically relevant surface densities and when fusion occurs between the fluorescent donor and unlabeled acceptor vesicles the fluorescent intensity can reflect the dynamics of lipid fusion.
To test this prediction, we first utilized computational methods to predict conserved fRNA secondary structures within multiple species alignments of Saccharomyces sensu strico genomes.
In the prediction of gene function we utilized human-reviewed Gene Ontology (GO) Biological Process (BP) codes, as available from Bioconductor [36], [37].
