AICM: A Genuine Framework for Correcting Inconsistency Between Large Pharmacogenomics Datasets

Abstract

The inconsistency of open pharmacogenomics datasets produced by different studies limits the usage of such datasets in many tasks, such as biomarker discovery. Investigation of multiple pharmacogenomics datasets confirmed that the pairwise sensitivity data correlation between drugs, or rows, across different studies (drug-wise) is relatively low, while the pairwise sensitivity data correlation between cell-lines, or columns, across different studies (cell-wise) is considerably strong. This common interesting observation across multiple pharmacogenomics datasets suggests the existence of subtle consistency among the different studies (i.e., strong cell-wise correlation). However, significant noises are also shown (i.e., weak drug-wise correlation) and have prevented researchers from comfortably using the data directly. Motivated by this observation, we propose a novel framework for addressing the inconsistency between large-scale pharmacogenomics data sets. Our method can significantly boost the drug-wise correlation and can be easily applied to resummarized and normalized datasets proposed by others. We also investigate our algorithm based on many different criteria to demonstrate that the corrected datasets are not only consistent, but also biologically meaningful. Eventually, we propose to extend our main algorithm into a framework, so that in the future when more datasets become publicly available, our framework can hopefully offer a "ground-truth" guidance for references.

TransCell: In silico characterization of genomic landscape and cellular responses from gene expressions through a two-step deep transfer learning

“…The inferior performance of perturbation experiments (CRISPR knockout and drug treatments) confirms the complicated cellular responses that could not be easily captured by linear models. Besides, the quality of the high throughput experiments sets back model development [43]. TransCell adequately addresses such challenges.…”

Precision Medicine: Improving health through high-resolution analysis of personal data

“…Another project helping to manage data related to precision medicine is the work of Zhiyue Tom Hu and colleagues [3], where they describe a framework for addressing inconsistency in large pharmacogenomic data sets, where individual potential therapeutics are screened against cancer cell lines. The method, Alternating Imputation and Correction Method (AICM), uses shared overlap of a handful of tested medications to bring divergent datasets into alignment for comparison across the full span of data.…”

Leveraging multi-source to resolve inconsistency across pharmacogenomic datasets in drug sensitivity prediction

“…Hafner et al proposed GR50, a metric to summarize drug sensitivity which demonstrated better performance in assessing the effects of drugs in dividing cells [19]. Hu, Zhiyue Tom, et al developed a method called AICM to correct inconsistency between large pharmacogenomics datasets based on cell-wise correlations which also improves drugwise correlations [20].…”