A machine learning approach for somatic mutation discovery

Exploring the Consistency of the Quality Scores with Machine Learning for Next-Generation Sequencing Experiments

Background

Next-generation sequencing enables massively parallel processing, allowing lower cost than the other sequencing technologies. In the subsequent analysis with the NGS data, one of the major concerns is the reliability of variant calls. Although researchers can utilize raw quality scores of variant calling, they are forced to start the further analysis without any preevaluation of the quality scores.

Method

We presented a machine learning approach for estimating quality scores of variant calls derived from BWA+GATK. We analyzed correlations between the quality score and these annotations, specifying informative annotations which were used as features to predict variant quality scores. To test the predictive models, we simulated 24 paired-end Illumina sequencing reads with 30x coverage base. Also, twenty-four human genome sequencing reads resulting from Illumina paired-end sequencing with at least 30x coverage were secured from the Sequence Read Archive.

Results

Using BWA+GATK, VCFs were derived from simulated and real sequencing reads. We observed that the prediction models learned by RFR outperformed other algorithms in both simulated and real data. The quality scores of variant calls were highly predictable from informative features of GATK Annotation Modules in the simulated human genome VCF data (R2: 96.7%, 94.4%, and 89.8% for RFR, MLR, and NNR, respectively). The robustness of the proposed data-driven models was consistently maintained in the real human genome VCF data (R2: 97.8% and 96.5% for RFR and MLR, respectively).

The promise of toxicogenomics for genetic toxicology: past, present and future

Toxicogenomics, the application of genomics to toxicology, was described as 'a new era' for toxicology. Standard toxicity tests typically involve a number of short-term bioassays that are costly, time consuming, require large numbers of animals and generally focus on a single end point. Toxicogenomics was heralded as a way to improve the efficiency of toxicity testing by assessing gene regulation across the genome, allowing rapid classification of compounds based on characteristic expression profiles. Gene expression microarrays could measure and characterise genome-wide gene expression changes in a single study and while transcriptomic profiles that can discriminate between genotoxic and non-genotoxic carcinogens have been identified, challenges with the approach limited its application. As such, toxicogenomics did not transform the field of genetic toxicology in the way it was predicted. More recently, next generation sequencing (NGS) technologies have revolutionised genomics owing to the fact that hundreds of billions of base pairs can be sequenced simultaneously cheaper and quicker than traditional Sanger methods. In relation to genetic toxicology, and thousands of cancer genomes have been sequenced with single-base substitution mutational signatures identified, and mutation signatures have been identified following treatment of cells with known or suspected environmental carcinogens. RNAseq has been applied to detect transcriptional changes following treatment with genotoxins; modified RNAseq protocols have been developed to identify adducts in the genome and Duplex sequencing is an example of a technique that has recently been developed to accurately detect mutation. Machine learning, including MutationSeq and SomaticSeq, has also been applied to somatic mutation detection and improvements in automation and/or the application of machine learning algorithms may allow high-throughput mutation sequencing in the future. This review will discuss the initial promise of transcriptomics for genetic toxicology, and how the development of NGS technologies and new machine learning algorithms may finally realise that promise.

Establishing guidelines to harmonize tumor mutational burden (TMB): in silico assessment of variation in TMB quantification across diagnostic platforms: phase I of the Friends of Cancer Research TMB Harmonization Project

BACKGROUND

Tumor mutational burden (TMB), defined as the number of somatic mutations per megabase of interrogated genomic sequence, demonstrates predictive biomarker potential for the identification of patients with cancer most likely to respond to immune checkpoint inhibitors. TMB is optimally calculated by whole exome sequencing (WES), but next-generation sequencing targeted panels provide TMB estimates in a time-effective and cost-effective manner. However, differences in panel size and gene coverage, in addition to the underlying bioinformatics pipelines, are known drivers of variability in TMB estimates across laboratories. By directly comparing panel-based TMB estimates from participating laboratories, this study aims to characterize the theoretical variability of panel-based TMB estimates, and provides guidelines on TMB reporting, analytic validation requirements and reference standard alignment in order to maintain consistency of TMB estimation across platforms.

METHODS

Eleven laboratories used WES data from The Cancer Genome Atlas Multi-Center Mutation calling in Multiple Cancers (MC3) samples and calculated TMB from the subset of the exome restricted to the genes covered by their targeted panel using their own bioinformatics pipeline (panel TMB). A reference TMB value was calculated from the entire exome using a uniform bioinformatics pipeline all members agreed on (WES TMB). Linear regression analyses were performed to investigate the relationship between WES and panel TMB for all 32 cancer types combined and separately. Variability in panel TMB values at various WES TMB values was also quantified using 95% prediction limits.

RESULTS

Study results demonstrated that variability within and between panel TMB values increases as the WES TMB values increase. For each panel, prediction limits based on linear regression analyses that modeled panel TMB as a function of WES TMB were calculated and found to approximately capture the intended 95% of observed panel TMB values. Certain cancer types, such as uterine, bladder and colon cancers exhibited greater variability in panel TMB values, compared with lung and head and neck cancers.

CONCLUSIONS

Increasing uptake of TMB as a predictive biomarker in the clinic creates an urgent need to bring stakeholders together to agree on the harmonization of key aspects of panel-based TMB estimation, such as the standardization of TMB reporting, standardization of analytical validation studies and the alignment of panel-based TMB values with a reference standard. These harmonization efforts should improve consistency and reliability of panel TMB estimates and aid in clinical decision-making.

Multimodal genomic features predict outcome of immune checkpoint blockade in non-small-cell lung cancer

Despite progress in immunotherapy, identifying patients that respond has remained a challenge. Through analysis of whole-exome and targeted sequence data from 5,449 tumors, we found a significant correlation between tumor mutation burden (TMB) and tumor purity, suggesting that low tumor purity tumors are likely to have inaccurate TMB estimates. We developed a new method to estimate a corrected TMB (cTMB) that was adjusted for tumor purity and more accurately predicted outcome to immune checkpoint blockade (ICB). To identify improved predictive markers together with cTMB, we performed whole-exome sequencing for 104 lung tumors treated with ICB. Through comprehensive analyses of sequence and structural alterations, we discovered a significant enrichment in activating mutations in receptor tyrosine kinase (RTK) genes in nonresponding tumors in three immunotherapy treated cohorts. An integrated multivariable model incorporating cTMB, RTK mutations, smoking-related mutational signature and human leukocyte antigen status provided an improved predictor of response to immunotherapy that was independently validated.

Fast-developing machine learning support complex system research in environmental chemistry

Machine learning will radically accelerate analysis of complex material networks in environmental chemistry.

FIREVAT: finding reliable variants without artifacts in human cancer samples using etiologically relevant mutational signatures

Background

Accurate identification of real somatic variants is a primary part of cancer genome studies and precision oncology. However, artifacts introduced in various steps of sequencing obfuscate confidence in variant calling. Current computational approaches to variant filtering involve intensive interrogation of Binary Alignment Map (BAM) files and require massive computing power, data storage, and manual labor. Recently, mutational signatures associated with sequencing artifacts have been extracted by the Pan-cancer Analysis of Whole Genomes (PCAWG) study. These spectrums can be used to evaluate refinement quality of a given set of somatic mutations.

Results

Here we introduce a novel variant refinement software, FIREVAT (FInding REliable Variants without ArTifacts), which uses known spectrums of sequencing artifacts extracted from one of the largest publicly available catalogs of human tumor samples. FIREVAT performs a quick and efficient variant refinement that accurately removes artifacts and greatly improves the precision and specificity of somatic calls. We validated FIREVAT refinement performance using orthogonal sequencing datasets totaling 384 tumor samples with respect to ground truth. Our novel method achieved the highest level of performance compared to existing filtering approaches. Application of FIREVAT on additional 308 The Cancer Genome Atlas (TCGA) samples demonstrated that FIREVAT refinement leads to identification of more biologically and clinically relevant mutational signatures as well as enrichment of sequence contexts associated with experimental errors. FIREVAT only requires a Variant Call Format file (VCF) and generates a comprehensive report of the variant refinement processes and outcomes for the user.

Conclusions

In summary, FIREVAT facilitates a novel refinement strategy using mutational signatures to distinguish artifactual point mutations called in human cancer samples. We anticipate that FIREVAT results will further contribute to precision oncology efforts that rely on accurate identification of variants, especially in the context of analyzing mutational signatures that bear prognostic and therapeutic significance. FIREVAT is freely available at https://github.com/cgab-ncc/FIREVAT

Recurrent PTPRT/JAK2 mutations in lung adenocarcinoma among African Americans

Reducing or eliminating persistent disparities in lung cancer incidence and survival has been challenging because our current understanding of lung cancer biology is derived primarily from populations of European descent. Here we show results from a targeted sequencing panel using NCI-MD Case Control Study patient samples and reveal a significantly higher prevalence of PTPRT and JAK2 mutations in lung adenocarcinomas among African Americans compared with European Americans. This increase in mutation frequency was validated with independent WES data from the NCI-MD Case Control Study and TCGA. We find that patients carrying these mutations have a concomitant increase in IL-6/STAT3 signaling and miR-21 expression. Together, these findings suggest the identification of these potentially actionable mutations could have clinical significance for targeted therapy and the enrollment of minority populations in clinical trials.

Artificial intelligence for precision medicine in neurodevelopmental disorders

The ambition of precision medicine is to design and optimize the pathway for diagnosis, therapeutic intervention, and prognosis by using large multidimensional biological datasets that capture individual variability in genes, function and environment. This offers clinicians the opportunity to more carefully tailor early interventions- whether treatment or preventative in nature-to each individual patient. Taking advantage of high performance computer capabilities, artificial intelligence (AI) algorithms can now achieve reasonable success in predicting risk in certain cancers and cardiovascular disease from available multidimensional clinical and biological data. In contrast, less progress has been made with the neurodevelopmental disorders, which include intellectual disability (ID), autism spectrum disorder (ASD), epilepsy and broader neurodevelopmental disorders. Much hope is pinned on the opportunity to quantify risk from patterns of genomic variation, including the functional characterization of genes and variants, but this ambition is confounded by phenotypic and etiologic heterogeneity, along with the rare and variable penetrant nature of the underlying risk variants identified so far. Structural and functional brain imaging and neuropsychological and neurophysiological markers may provide further dimensionality, but often require more development to achieve sensitivity for diagnosis. Herein, therefore, lies a precision medicine conundrum: can artificial intelligence offer a breakthrough in predicting risks and prognosis for neurodevelopmental disorders? In this review we will examine these complexities, and consider some of the strategies whereby artificial intelligence may overcome them.

Somatic mutations - Evolution within the individual

With the rapid advancement of sequencing technologies over the last two decades, it is becoming feasible to detect rare variants from somatic tissue samples. Studying such somatic mutations can provide deep insights into various senescence-related diseases, including cancer, inflammation, and sporadic psychiatric disorders. While it is still a difficult task to identify true somatic mutations, relentless efforts to combine experimental and computational methods have made it possible to obtain reliable data. Furthermore, state-of-the-art machine learning approaches have drastically improved the efficiency and sensitivity of these methods. Meanwhile, we can regard somatic mutations as a counterpart of germline mutations, and it is possible to apply well-formulated mathematical frameworks developed for population genetics and molecular evolution to analyze this 'somatic evolution'. For example, retrospective cell lineage tracing is a promising technique to elucidate the mechanism of pre-diseases using single-cell RNA-sequencing (scRNA-seq) data.

Integration of Random Forest Classifiers and Deep Convolutional Neural Networks for Classification and Biomolecular Modeling of Cancer Driver Mutations

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.

Translating cancer genomics into precision medicine with artificial intelligence: applications, challenges and future perspectives

In the field of cancer genomics, the broad availability of genetic information offered by next-generation sequencing technologies and rapid growth in biomedical publication has led to the advent of the big-data era. Integration of artificial intelligence (AI) approaches such as machine learning, deep learning, and natural language processing (NLP) to tackle the challenges of scalability and high dimensionality of data and to transform big data into clinically actionable knowledge is expanding and becoming the foundation of precision medicine. In this paper, we review the current status and future directions of AI application in cancer genomics within the context of workflows to integrate genomic analysis for precision cancer care. The existing solutions of AI and their limitations in cancer genetic testing and diagnostics such as variant calling and interpretation are critically analyzed. Publicly available tools or algorithms for key NLP technologies in the literature mining for evidence-based clinical recommendations are reviewed and compared. In addition, the present paper highlights the challenges to AI adoption in digital healthcare with regard to data requirements, algorithmic transparency, reproducibility, and real-world assessment, and discusses the importance of preparing patients and physicians for modern digitized healthcare. We believe that AI will remain the main driver to healthcare transformation toward precision medicine, yet the unprecedented challenges posed should be addressed to ensure safety and beneficial impact to healthcare.

Dynamics of Tumor and Immune Responses during Immune Checkpoint Blockade in Non-Small Cell Lung Cancer

Despite the initial successes of immunotherapy, there is an urgent clinical need for molecular assays that identify patients more likely to respond. Here, we report that ultrasensitive measures of circulating tumor DNA (ctDNA) and T-cell expansion can be used to assess responses to immune checkpoint blockade in metastatic lung cancer patients (N = 24). Patients with clinical response to therapy had a complete reduction in ctDNA levels after initiation of therapy, whereas nonresponders had no significant changes or an increase in ctDNA levels. Patients with initial response followed by acquired resistance to therapy had an initial drop followed by recrudescence in ctDNA levels. Patients without a molecular response had shorter progression-free and overall survival compared with molecular responders [5.2 vs. 14.5 and 8.4 vs. 18.7 months; HR 5.36; 95% confidence interval (CI), 1.57-18.35; P = 0.007 and HR 6.91; 95% CI, 1.37-34.97; P = 0.02, respectively], which was detected on average 8.7 weeks earlier and was more predictive of clinical benefit than CT imaging. Expansion of T cells, measured through increases of T-cell receptor productive frequencies, mirrored ctDNA reduction in response to therapy. We validated this approach in an independent cohort of patients with early-stage non-small cell lung cancer (N = 14), where the therapeutic effect was measured by pathologic assessment of residual tumor after anti-PD1 therapy. Consistent with our initial findings, early ctDNA dynamics predicted pathologic response to immune checkpoint blockade. These analyses provide an approach for rapid determination of therapeutic outcomes for patients treated with immune checkpoint inhibitors and have important implications for the development of personalized immune targeted strategies.Significance: Rapid and sensitive detection of circulating tumor DNA dynamic changes and T-cell expansion can be used to guide immune targeted therapy for patients with lung cancer.See related commentary by Zou and Meyerson, p. 1038.