-
Notifications
You must be signed in to change notification settings - Fork 0
Home
Welcome to the DeepPhenotypingHPO wiki!
Next Generation Sequencing has lead to a rapid expansion of the human genetic pathology 'atlas'. However, the rapidly increasing group of rare genetic disorders remain poorly understood as NGS techniques proves incapable of differentiating polymorphisms from clinically relevant variants. Clinical adoption of these findings now often require highly time-expensive in-vitro experiments. We argue that you could acquire novel insights by employing phenotypical data to infer the functional impact and overall clinical relevancy of these genetic variants [1].
Deep Phenotyping is a comprehensive analysis based on the principle that clinical similarities are indicative of a shared underlying pathopysiology. This technique allows researchers to extrapolate pathophysiological insights from well-known diseases to rare disorders, rendering increased insight in underlying pathophysiology of rare disorders. However, there is no consensus on the best deep phenotyping strategy yet.
One major advantage of Deep Phenotyping is that it utilizes information that is already available: phenotypes. To combat the data scarcity, these phenotypic descriptions can directly be extracted from case studies. Hence, we build a HPO-extraction tool to facilitate deep phenotyping studies.
The hypothesis that similarities in phenotypes suggest a shared pathophysiology was explored in a paper by Haijes et al. in 2020.
Prioritization - To uncover the causative genes we used an established method called PHRANK. The code for this project is available at: https://bitbucket.org/bejerano/phrank/src/master/
-
HPO_clustering_Freq.ipynb: Python notebook file with clustering steps divided into modules -
HPO extraction tool.ipynb: Python notebook file with Case Study Extractor steps divided into modules -
Phrank.ipynb: Python notebook file for Causative Gene Identification analysis with Phrank -
src/*: All python scripts are stored here -
src/CaseStudyExtractor.py: Python script for the commandline runs Case Study Extractor pipeline -
src/DeepPhenotyping.py: Python script for the commandline runs Case Study Extractor pipeline -
src/DeepPhenotyping_functions.py: Python script that stores all the major functions for Case Study Extractor and Deep Phenotyping -
PhenoTool/*: Directory consisting of the different phenotools that were evaluated -
results/*: Case studies processed by the Case Study Extractor -
TSNE/*: All interactive TSNE plots are stored here -
PCA/*: All interactive PCA plots are stored here