Metadata-Version: 2.0
Name: deepsvr
Version: 0.0.1.post1
Summary: Automated Somatic Variant Refinement by Deep Learning
Home-page: https://github.com/griffithlab/manual_review_classifier/tree/master/deepsvr
Author: The Griffith Lab
Author-email: obigriffith@wustl.edu
License: UNKNOWN
Description-Content-Type: UNKNOWN
Keywords: somatic variant refinement manual review deep learning genomics bioinformatics sequencing
Platform: UNKNOWN
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Science/Research
Classifier: Topic :: Scientific/Engineering :: Bio-Informatics
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3.6
Requires-Python: >=3
Requires-Dist: click
Requires-Dist: convert-zero-one-based
Requires-Dist: h5py
Requires-Dist: keras
Requires-Dist: matplotlib
Requires-Dist: numpy
Requires-Dist: pandas
Requires-Dist: seaborn
Requires-Dist: setuptools
Requires-Dist: sklearn
Requires-Dist: tensorflow

Cancer genomic analysis requires accurate identification of somatic variants in sequencing data. Manual review to refine somatic variant calls is required as a final step after automated processing. However, manual variant refinement is time-consuming, costly, poorly standardized, and non-reproducible. Here, we systematized and standardized somatic variant refinement using a machine learning approach. The final model accurately recapitulated manual somatic variant refinement labels using an independent testing set and accurately predicted somatic variants confirmed by orthogonal validation sequencing data. The model improves on manual somatic variant refinement by reducing bias on calls usually subject to high inter-reviewer variability.


