Dataset Information

Integration of intra-sample contextual error modeling for improved detection of somatic mutations from deep sequencing.

ABSTRACT: Sensitive mutation detection by next-generation sequencing is critical for early cancer detection, monitoring minimal/measurable residual disease (MRD), and guiding precision oncology. Nevertheless, because of artifacts introduced during library preparation and sequencing, the detection of low-frequency variants at high specificity is problematic. Here, we present Espresso, an error suppression method that considers local sequence features to accurately detect single-nucleotide variants (SNVs). Compared to other advanced error suppression techniques, Espresso consistently demonstrated lower numbers of false-positive mutation calls and greater sensitivity. We demonstrated Espresso's superior performance in detecting MRD in the peripheral blood of patients with acute myeloid leukemia (AML) throughout their treatment course. Furthermore, we showed that accurate mutation calling in a small number of informative genomic loci might provide a cost-efficient strategy for pragmatic risk prediction of AML development in healthy individuals. More broadly, we aim for Espresso to aid with accurate mutation detection in many other research and clinical settings.

SUBMITTER: Abelson S

PROVIDER: S-EPMC7725472 | biostudies-literature | 2020 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Integration of intra-sample contextual error modeling for improved detection of somatic mutations from deep sequencing.

Abelson Sagi S Zeng Andy G X AGX Nofech-Mozes Ido I Wang Ting Ting TT Ng Stanley W K SWK Minden Mark D MD Pugh Trevor J TJ Awadalla Philip P Shlush Liran I LI Murphy Tracy T Chan Steven M SM Dick John E JE Bratman Scott V SV

Science advances 20201209 50

Sensitive mutation detection by next-generation sequencing is critical for early cancer detection, monitoring minimal/measurable residual disease (MRD), and guiding precision oncology. Nevertheless, because of artifacts introduced during library preparation and sequencing, the detection of low-frequency variants at high specificity is problematic. Here, we present Espresso, an error suppression method that considers local sequence features to accurately detect single-nucleotide variants (SNVs). ...[more]

PMID: 33298453

Similar Datasets

Project description:Development of machine learning solutions for prediction of functional and clinical significance of cancer driver genes and mutations are paramount in modern biomedical research and have gained a significant momentum in a recent decade. In this work, we integrate different machine learning approaches, including tree based methods, random forest and gradient boosted tree (GBT) classifiers along with deep convolutional neural networks (CNN) for prediction of cancer driver mutations in the genomic datasets. The feasibility of CNN in using raw nucleotide sequences for classification of cancer driver mutations was initially explored by employing label encoding, one hot encoding, and embedding to preprocess the DNA information. These classifiers were benchmarked against their tree-based alternatives in order to evaluate the performance on a relative scale. We then integrated DNA-based scores generated by CNN with various categories of conservational, evolutionary and functional features into a generalized random forest classifier. The results of this study have demonstrated that CNN can learn high level features from genomic information that are complementary to the ensemble-based predictors often employed for classification of cancer mutations. By combining deep learning-generated score with only two main ensemble-based functional features, we can achieve a superior performance of various machine learning classifiers. Our findings have also suggested that synergy of nucleotide-based deep learning scores and integrated metrics derived from protein sequence conservation scores can allow for robust classification of cancer driver mutations with a limited number of highly informative features. Machine learning predictions are leveraged in molecular simulations, protein stability, and network-based analysis of cancer mutations in the protein kinase genes to obtain insights about molecular signatures of driver mutations and enhance the interpretability of cancer-specific classification models.

Dataset Information

Integration of intra-sample contextual error modeling for improved detection of somatic mutations from deep sequencing.

Publications

Integration of intra-sample contextual error modeling for improved detection of somatic mutations from deep sequencing.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets