In silico comparative characterization of pharmacogenomic missense variants

A novel machine learning-based approach for the computational functional assessment of pharmacogenomic variants

BACKGROUND

The field of pharmacogenomics focuses on the way a person's genome affects his or her response to a certain dose of a specified medication. The main aim is to utilize this information to guide and personalize the treatment in a way that maximizes the clinical benefits and minimizes the risks for the patients, thus fulfilling the promises of personalized medicine. Technological advances in genome sequencing, combined with the development of improved computational methods for the efficient analysis of the huge amount of generated data, have allowed the fast and inexpensive sequencing of a patient's genome, hence rendering its incorporation into clinical routine practice a realistic possibility.

METHODS

This study exploited thoroughly characterized in functional level SNVs within genes involved in drug metabolism and transport, to train a classifier that would categorize novel variants according to their expected effect on protein functionality. This categorization is based on the available in silico prediction and/or conservation scores, which are selected with the use of recursive feature elimination process. Toward this end, information regarding 190 pharmacovariants was leveraged, alongside with 4 machine learning algorithms, namely AdaBoost, XGBoost, multinomial logistic regression, and random forest, of which the performance was assessed through 5-fold cross validation.

RESULTS

All models achieved similar performance toward making informed conclusions, with RF model achieving the highest accuracy (85%, 95% CI: 0.79, 0.90), as well as improved overall performance (precision 85%, sensitivity 84%, specificity 94%) and being used for subsequent analyses. When applied on real world WGS data, the selected RF model identified 2 missense variants, expected to lead to decreased function proteins and 1 to increased. As expected, a greater number of variants were highlighted when the approach was used on NGS data derived from targeted resequencing of coding regions. Specifically, 71 variants (out of 156 with sufficient annotation information) were classified as to "Decreased function," 41 variants as "No" function proteins, and 1 variant in "Increased function."

CONCLUSION

Overall, the proposed RF-based classification model holds promise to lead to an extremely useful variant prioritization and act as a scoring tool with interesting clinical applications in the fields of pharmacogenomics and personalized medicine.

Global spectrum of population-specific common missense variation in cytochrome P450 pharmacogenes

Next-generation sequencing technology has afforded the discovery of many novel variants that are of significance to inheritable pharmacogenomics (PGx) traits but a large proportion of them have unknown consequences. These include missense variants resulting in single amino acid substitutions in cytochrome P450 (CYP) proteins that can impair enzyme function, leading to altered drug efficacy and toxicity. While most unknown variants are rare, an overlooked minority are variants that are collectively rare but enriched in specific populations. Here, we analyzed sequence variation data in 141,456 individuals from across eight study populations in gnomAD for 38 CYP genes to identify such variants in addition to common variants. By further comparison with data from two PGx-specific databases (PharmVar and PharmGKB) and ClinVar, we identified 234 missense variants in 35 CYP genes, of which 107 were unknown to these databases. Most unknown variants (n = 83) were population-specific common variants and several (n = 7) were found in important CYP pharmacogenes (CYP2D6, CYP4F2, and CYP2C19). Overall, 29% (n = 31) of 107 unknown variants were predicted to affect CYP enzyme function although further biochemical characterization is necessary. These variants may elucidate part of the unexplained interpopulation differences observed in drug response.

Common Treatment, Common Variant: Evolutionary Prediction of Functional Pharmacogenomic Variants

Pharmacogenomics holds the promise of personalized drug efficacy optimization and drug toxicity minimization. Much of the research conducted to date, however, suffers from an ascertainment bias towards European participants. Here, we leverage publicly available, whole genome sequencing data collected from global populations, evolutionary characteristics, and annotated protein features to construct a new in silico machine learning pharmacogenetic identification method called XGB-PGX. When applied to pharmacogenetic data, XGB-PGX outperformed all existing prediction methods and identified over 2000 new pharmacogenetic variants. While there are modest pharmacogenetic allele frequency distribution differences across global population samples, the most striking distinction is between the relatively rare putatively neutral pharmacogene variants and the relatively common established and newly predicted functional pharamacogenetic variants. Our findings therefore support a focus on individual patient pharmacogenetic testing rather than on clinical presumptions about patient race, ethnicity, or ancestral geographic residence. We further encourage more attention be given to the impact of common variation on drug response and propose a new ‘common treatment, common variant’ perspective for pharmacogenetic prediction that is distinct from the types of variation that underlie complex and Mendelian disease. XGB-PGX has identified many new pharmacovariants that are present across all global communities; however, communities that have been underrepresented in genomic research are likely to benefit the most from XGB-PGX’s in silico predictions.

Technologies for Pharmacogenomics: A Review

The continuous development of new genotyping technologies requires awareness of their potential advantages and limitations concerning utility for pharmacogenomics (PGx). In this review, we provide an overview of technologies that can be applied in PGx research and clinical practice. Most commonly used are single nucleotide variant (SNV) panels which contain a pre-selected panel of genetic variants. SNV panels offer a short turnaround time and straightforward interpretation, making them suitable for clinical practice. However, they are limited in their ability to assess rare and structural variants. Next-generation sequencing (NGS) and long-read sequencing are promising technologies for the field of PGx research. Both NGS and long-read sequencing often provide more data and more options with regard to deciphering structural and rare variants compared to SNV panels-in particular, in regard to the number of variants that can be identified, as well as the option for haplotype phasing. Nonetheless, while useful for research, not all sequencing data can be applied to clinical practice yet. Ultimately, selecting the right technology is not a matter of fact but a matter of choosing the right technique for the right problem.

Development of the PGx-Passport: A Panel of Actionable Germline Genetic Variants for Pre-Emptive Pharmacogenetic Testing

Pre-emptive pharmacogenetics (PGx) testing of a panel of germline genetic variants represents a new model for personalized medicine. Clinical impact of PGx testing is maximized when all variant alleles for which actionable clinical guidelines are available are included in the test panel. However, no such standardized panel has been presented to date, impeding adoption, exchange, and continuity of PGx testing. We, therefore, developed such a panel, hereafter called the PGx-Passport, based on the actionable Dutch Pharmacogenetics Working Group (DPWG) guidelines. Germline-variant alleles were systematically selected using predefined criteria regarding allele population frequencies, effect on protein functionality, and association with drug response. A PGx-Passport of 58 germline variant alleles, located within 14 genes (CYP2B6, CYP2C9, CYP2C19, CYP2D6, CYP3A5, DPYD, F5, HLA-A, HLA-B, NUDT15, SLCO1B1, TPMT, UGT1A1, and VKORC1) was composed. This PGx-Passport can be used in combination with the DPWG guidelines to optimize drug prescribing for 49 commonly prescribed drugs.

Exploring timing activation of functional pathway based on differential co-expression analysis in preimplantation embryogenesis

Recent genome-wide omics studies have confirmed the early embryogenesis strictly dependent on the rigorous spatiotemporal activation and multilevel regulation. However, the full effect of functional pathway was not considered. To obtain complete understanding of the gene activation during early development, we performed systematic comparisons based on differential co-expression analysis for bovine preimplantation embryo development (PED). The results confirmed that the functional pathways actively transcribes as early as the 2-cell and 4-cell waves, which Basal transcription factor, Endocytosis and Spliceosome pathway can represent first signs of embryonic activity. Endocytosis act as one of master activators for uncovering a series of successive waves of maternal pioneer signal regulator with the help of Spliceosome complex. Furthermore, the results showed that pattern recognition receptors began to perform its essential function at 4-cell stage, which might be needed to coordinate the later major activation. And finally, our work presented a probable dynamic landscape of key functional pathways for embryogenesis. A clearer understanding of early embryo development will be helpful for Assisted Reproductive Technology (ART) and Regenerative Medicine (RM).

Computational methods and resources for the interpretation of genomic variants in cancer

The recent improvement of the high-throughput sequencing technologies is having a strong impact on the detection of genetic variations associated with cancer. Several institutions worldwide have been sequencing the whole exomes and or genomes of cancer patients in the thousands, thereby providing an invaluable collection of new somatic mutations in different cancer types. These initiatives promoted the development of methods and tools for the analysis of cancer genomes that are aimed at studying the relationship between genotype and phenotype in cancer. In this article we review the online resources and computational tools for the analysis of cancer genome. First, we describe the available repositories of cancer genome data. Next, we provide an overview of the methods for the detection of genetic variation and computational tools for the prioritization of cancer related genes and causative somatic variations. Finally, we discuss the future perspectives in cancer genomics focusing on the impact of computational methods and quantitative approaches for defining personalized strategies to improve the diagnosis and treatment of cancer.

Progress towards the integration of pharmacogenomics in practice

Understanding the role genes and genetic variants play in clinical treatment response continues to be an active area of research with the goal of common clinical use. This goal has developed into today's industry of pharmacogenomics, where new drug-gene relationships are discovered and further characterized, published and then curated into national and international resources for use by researchers and clinicians. These efforts have given us insight into what a pharmacogenomic variant is, and how it differs from human disease variants and common polymorphisms. While publications continue to reveal pharmacogenomic relationships between genes and specific classes of drugs, many challenges remain toward the goal of widespread use clinically. First, the clinical guidelines for pharmacogenomic testing are still in their infancy. Second, sequencing technologies are changing rapidly making it somewhat unclear what genetic data will be available to the clinician at the time of care. Finally, what and when to return data to a patient is an area under constant debate. New innovations such as PheWAS approaches and whole genome sequencing studies are enabling a tsunami of new findings. In this review, pharmacogenomic variants, pharmacogenomic resources, interpretation clinical guidelines and challenges, such as WGS approaches, and the impact of pharmacogenomics on drug development and regulatory approval are reviewed.