Deep Phenotyping - Harnessing Data Richness for Unsupervised High-Content Analysis
Poster May 05, 2016
Huang Dong, Wang Yi, Maciej Hermanowicz, Ke Yiping, Maja Choma, Lee Kee Khoon, Frederic Bard
High-content analysis (HCA) has recently arisen as a novel systematic approach to cell biology research. Due to the intrinsically high variability and noise level of the assay, it has been primarily applied to supervised analysis, where the phenotypes are known beforehand and manual labels are provided to guide the analysis. Furthermore, most existing research uses only a few parameters, failing to exploit the richness of the raw images.
In this work, we develop an end-to-end computational framework focused on unsupervised HCA. The framework can identify unknown subtle phenotypes from massive multi-parametric data, thus dubbed “Deep Phenotyping”, and correlate them with gene perturbations. It consists of a series of modules: high-dimensional parameter extraction for each cell from raw images, multi-level quality control using both supervised and unsupervised methods with minimal labeling effort, a novelty detection algorithm to automatically retrieve cells that differ from the control, and an unsupervised clustering algorithm to identify unknown phenotypes in the multi-parametric space.
The framework is modularized and applicable to a variety of HCA tasks. We demonstrate its use on the phenotypic analysis of the Golgi apparatus. Novel Golgi phenotypes beyond the three known ones are identified, and potential protein-protein interactions regulating Golgi functions are revealed.
Screening of ALDH2 Interactome by Combined Immunoprecipitation and Mass SpectrometryPoster
ALDH2 may have an unidentified protein interaction network, which would help elucidate its importance in various cellular functions. The combination of immunoprecipitation and mass spectrometry will give us insights that could not be easily attained otherwise.READ MORE
Identification of genes involved in cell cycle regulation using arrayed synthetic CRISPR RNA libraries in a multiparameter high-content assayPoster
Gene knockout using CRISPR-Cas9 has dramatically transformed biological research and has been rapidly applied to loss-of-function screening primarily using pooled lentiviral sgRNA libraries. A synthetic CRISPR RNA (crRNA) approach is amenable to screening in arrayed, well-by-well fashion and expands the types of phenotypic readouts that can be used, including high-content and morphology-based assays.READ MORE
Nitrogen metabolism in Mycobacterium tuberculosis: a systems-based approachPoster
Previous studies suggested that Mycobacterium tuberculosis obtains nitrogen from a diverse range of intracellular nutrients including amino acids. Here, we use a novel system’s based three-pronged approach to define pathways for uptake and assimilation of nitrogen.READ MORE