We then investigated whether phenotype 1/2 or cells based on the FAB AML classification were associated with a particular phospho-protein expression pattern.
In summary, this study constitutes an "atlas" of histone PTMs across the reported lines, and using this atlas, one may potentially attain a novel molecular classification of cancer cells based on their chromatin modification landscape.
Integrating the results of threshold-based and the cell-based classification, we found that most of the false-positive images given by threshold-based classification are now correctly classified, with only one false positive (0.4%).
We also provided cell classification based on cell cycle states and analyzed co-expression network modules of HeLa S3 cells.
And, at the cellular level, Xu et al. target the identification of antinuclear antibodies as evidence of autoimmune diseases with their work "HEp-2 Cells Classification Based on a Linear Local Distance Coding Framework".
This analytical theory allows to quickly estimate the parameters for optimal training of the distributed cell population based classifier for a given classification problem.
Strikingly, the robustness of the cell line classifications based on each drug response metric could be improved by using the TAS transformation (Fig. 4A), most significantly with the IC50 metric (P=0.009, Wilcoxon rank-sum test).
It is the activity of the chromatin modifying enzymes that is most informative for designing an accurate classification of cell lines based on their chromatin states.
Using cancer cell line profiling data sets, we constructed a functional classification across 107 cancer cell models, based on their common and unique target addiction signatures.
Information on cell proliferation may be a useful adjunct to histologically based tumour classification in the understanding of tumour behaviour.
We first derived unsupervised molecular and functional classifications of the 107 cancer cell lines based on their transcriptional, drug response and target addiction profiles.
