Bedside Back to Bench: Building Bridges between Basic and Clinical Genomic Research.

Genome sequencing has revolutionized the diagnosis of genetic diseases. Close collaborations between basic scientists and clinical genomicists are now needed to link genetic variants with disease causation. To facilitate such collaborations, we recommend prioritizing clinically relevant genes for functional studies, developing reference variant-phenotype databases, adopting phenotype description standards, and promoting data sharing.

Will variants of uncertain significance still exist in 2030?

In 2020, the National Human Genome Research Institute (NHGRI) made ten "bold predictions," including that "the clinical relevance of all encountered genomic variants will be readily predictable, rendering the diagnostic designation 'variant of uncertain significance (VUS)' obsolete." We discuss the prospects for this prediction, arguing that many, if not most, VUS in coding regions will be resolved by 2030. We outline a confluence of recent changes making this possible, especially advances in the standards for variant classification that better leverage diverse types of evidence, improvements in computational variant effect predictor performance, scalable multiplexed assays of variant effect capable of saturating the genome, and data-sharing efforts that will maximize the information gained from each new individual sequenced and variant interpreted. We suggest that clinicians and researchers can realize a future where VUSs have largely been eliminated, in line with the NHGRI's bold prediction. The length of time taken to reach this future, and thus whether we are able to achieve the goal of largely eliminating VUSs by 2030, is largely a consequence of the choices made now and in the next few years. We believe that investing in eliminating VUSs is worthwhile, since their predominance remains one of the biggest challenges to precision genomic medicine.

Calibration of variant effect predictors on genome-wide data masks heterogeneous performance across genes.

In silico variant effect predictions are available for nearly all missense variants but played a minimal role in clinical variant classification because they were deemed to provide only supporting evidence. Recently, the ClinGen Sequence Variant Interpretation (SVI) Working Group updated recommendations for variant effect prediction use. By analyzing control pathogenic and benign variants across all genes, they were able to compute evidence strength for predictor score intervals with some intervals generating moderate, strong, or even very strong evidence. However, this genome-wide approach could obscure heterogeneous predictor performance in different genes. We quantified the gene-by-gene performance of two top predictors, REVEL and BayesDel, by analyzing control variants in each predictor score interval in 3,668 disease-relevant genes. Approximately 10% of intervals had sufficient control variants for analysis, and ∼70% of these intervals exceeded the maximum number of incorrect predictions implied by the SVI recommendations. These trending discordant intervals arose owing to the divergence of the gene-specific distribution of predictions from the genome-wide distribution, suggesting that gene-specific calibration is needed in many cases. Approximately 22% of ClinVar missense variants of uncertain significance in genes we analyzed (REVEL = 100,629, BayesDel = 71,928) had predictions in trending discordant intervals. Thus, genome-wide calibrations could result in many variants receiving inappropriate evidence strength. To facilitate a review of the SVI's calibrations, we developed a web application enabling visualization of gene-specific predictions and trending concordant and discordant intervals.

Multiplexed profiling of intracellular protein abundance, activity, interactions and druggability with LABEL-seq.

Here we describe labeling with barcodes and enrichment for biochemical analysis by sequencing (LABEL-seq), an assay for massively parallel profiling of pooled protein variants in human cells. By leveraging the intracellular self-assembly of an RNA-binding domain (RBD) with a stable, variant-encoding RNA barcode, LABEL-seq facilitates the direct measurement of protein properties and functions using simple affinity enrichments of RBD protein fusions, followed by high-throughput sequencing of co-enriched barcodes. Measurement of ~20,000 variant effects for ~1,600 BRaf variants revealed that variation at positions frequently mutated in cancer minimally impacted intracellular abundance but could dramatically alter activity, protein-protein interactions and druggability. Integrative analysis identified networks of positions with similar biochemical roles and enabled modeling of variant effects on cell proliferation and small molecule-promoted degradation. Thus, LABEL-seq enables direct measurement of multiple biochemical properties in a native cellular context, providing insights into protein function, disease mechanisms and druggability.

A Combined Approach Reveals a Regulatory Mechanism Coupling Src's Kinase Activity, Localization, and Phosphotransferase-Independent Functions.

Multiple layers of regulation modulate the activity and localization of protein kinases. However, many details of kinase regulation remain incompletely understood. Here, we apply saturation mutagenesis and a chemical genetic method for allosterically modulating kinase global conformation to Src kinase, providing insight into known regulatory mechanisms and revealing a previously undiscovered interaction between Src's SH4 and catalytic domains. Abrogation of this interaction increased phosphotransferase activity, promoted membrane association, and provoked phosphotransferase-independent alterations in cell morphology. Thus, Src's SH4 domain serves as an intramolecular regulator coupling catalytic activity, global conformation, and localization, as well as mediating a phosphotransferase-independent function. Sequence conservation suggests that the SH4 domain regulatory interaction exists in other Src-family kinases. Our combined approach's ability to reveal a regulatory mechanism in one of the best-studied kinases suggests that it could be applied broadly to provide insight into kinase structure, regulation, and function.

Antigen perception in T cells by long-term Erk and NFAT signaling dynamics.

Immune system threat detection hinges on T cells' ability to perceive varying peptide-major histocompatibility complex (pMHC) antigens. As the Erk and NFAT pathways link T cell receptor engagement to gene regulation, their signaling dynamics may convey information about pMHC inputs. To test this idea, we developed a dual reporter mouse strain and a quantitative imaging assay that, together, enable simultaneous monitoring of Erk and NFAT dynamics in live T cells over day-long timescales as they respond to varying pMHC inputs. Both pathways initially activate uniformly across various pMHC inputs but diverge only over longer (9+ h) timescales, enabling independent encoding of pMHC affinity and dose. These late signaling dynamics are decoded via multiple temporal and combinatorial mechanisms to generate pMHC-specific transcriptional responses. Our findings underscore the importance of long timescale signaling dynamics in antigen perception and establish a framework for understanding T cell responses under diverse contexts.

Measuring Pharmacogene Variant Function at Scale Using Multiplexed Assays.

As costs of next-generation sequencing decrease, identification of genetic variants has far outpaced our ability to understand their functional consequences. This lack of understanding is a central challenge to a key promise of pharmacogenomics: using genetic information to guide drug selection and dosing. Recently developed multiplexed assays of variant effect enable experimental measurement of the function of thousands of variants simultaneously. Here, we describe multiplexed assays that have been performed on nearly 25,000 variants in eight key pharmacogenes (ADRB2, CYP2C9, CYP2C19, NUDT15, SLCO1B1, TMPT, VKORC1, and the LDLR promoter), discuss advances in experimental design, and explore key challenges that must be overcome to maximize the utility of multiplexed functional data.

Pacybara: accurate long-read sequencing for barcoded mutagenized allelic libraries.

MOTIVATION: Long-read sequencing technologies, an attractive solution for many applications, often suffer from higher error rates. Alignment of multiple reads can improve base-calling accuracy, but some applications, e.g. sequencing mutagenized libraries where multiple distinct clones differ by one or few variants, require the use of barcodes or unique molecular identifiers. Unfortunately, sequencing errors can interfere with correct barcode identification, and a given barcode sequence may be linked to multiple independent clones within a given library. RESULTS: Here we focus on the target application of sequencing mutagenized libraries in the context of multiplexed assays of variant effects (MAVEs). MAVEs are increasingly used to create comprehensive genotype-phenotype maps that can aid clinical variant interpretation. Many MAVE methods use long-read sequencing of barcoded mutant libraries for accurate association of barcode with genotype. Existing long-read sequencing pipelines do not account for inaccurate sequencing or nonunique barcodes. Here, we describe Pacybara, which handles these issues by clustering long reads based on the similarities of (error-prone) barcodes while also detecting barcodes that have been associated with multiple genotypes. Pacybara also detects recombinant (chimeric) clones and reduces false positive indel calls. In three example applications, we show that Pacybara identifies and correctly resolves these issues. AVAILABILITY AND IMPLEMENTATION: Pacybara, freely available at https://github.com/rothlab/pacybara, is implemented using R, Python, and bash for Linux. It runs on GNU/Linux HPC clusters via Slurm, PBS, or GridEngine schedulers. A single-machine simplex version is also available.

A chemically controlled Cas9 switch enables temporal modulation of diverse effectors.

CRISPR-Cas9 has yielded a plethora of effectors, including targeted transcriptional activators, base editors and prime editors. Current approaches for inducibly modulating Cas9 activity lack temporal precision and require extensive screening and optimization. We describe a versatile, chemically controlled and rapidly activated single-component DNA-binding Cas9 switch, ciCas9, which we use to confer temporal control over seven Cas9 effectors, including two cytidine base editors, two adenine base editors, a dual base editor, a prime editor and a transcriptional activator. Using these temporally controlled effectors, we analyze base editing kinetics, showing that editing occurs within hours and that rapid early editing of nucleotides predicts eventual editing magnitude. We also reveal that editing at preferred nucleotides within target sites increases the frequency of bystander edits. Thus, the ciCas9 switch offers a simple, versatile approach to generating chemically controlled Cas9 effectors, informing future effector engineering and enabling precise temporal effector control for kinetic studies.

Massively parallel characterization of CYP2C9 variant enzyme activity and abundance.

CYP2C9 encodes a cytochrome P450 enzyme responsible for metabolizing up to 15% of small molecule drugs, and CYP2C9 variants can alter the safety and efficacy of these therapeutics. In particular, the anti-coagulant warfarin is prescribed to over 15 million people annually and polymorphisms in CYP2C9 can affect individual drug response and lead to an increased risk of hemorrhage. We developed click-seq, a pooled yeast-based activity assay, to test thousands of variants. Using click-seq, we measured the activity of 6,142 missense variants in yeast. We also measured the steady-state cellular abundance of 6,370 missense variants in a human cell line by using variant abundance by massively parallel sequencing (VAMP-seq). These data revealed that almost two-thirds of CYP2C9 variants showed decreased activity and that protein abundance accounted for half of the variation in CYP2C9 function. We also measured activity scores for 319 previously unannotated human variants, many of which may have clinical relevance.

Closing the gap: Systematic integration of multiplexed functional data resolves variants of uncertain significance in BRCA1, TP53, and PTEN.

Clinical interpretation of missense variants is challenging because the majority identified by genetic testing are rare and their functional effects are unknown. Consequently, most variants are of uncertain significance and cannot be used for clinical diagnosis or management. Although not much can be done to ameliorate variant rarity, multiplexed assays of variant effect (MAVEs), where thousands of single-nucleotide variant effects are simultaneously measured experimentally, provide functional evidence that can help resolve variants of unknown significance (VUSs). However, a rigorous assessment of the clinical value of multiplexed functional data for variant interpretation is lacking. Thus, we systematically combined previously published BRCA1, TP53, and PTEN multiplexed functional data with phenotype and family history data for 324 VUSs identified by a single diagnostic testing laboratory. We curated 49,281 variant functional scores from MAVEs for these three genes and integrated four different TP53 multiplexed functional datasets into a single functional prediction for each variant by using machine learning. We then determined the strength of evidence provided by each multiplexed functional dataset and reevaluated 324 VUSs. Multiplexed functional data were effective in driving variant reclassification when combined with clinical data, eliminating 49% of VUSs for BRCA1, 69% for TP53, and 15% for PTEN. Thus, multiplexed functional data, which are being generated for numerous genes, are poised to have a major impact on clinical variant interpretation.

Aggregating knowledge about prions and amyloid.

Prions, self-propagating protein structures that can be transmitted between cells and different organisms, usually consist of ordered protein aggregates. Alberti et al. (2009) now present a systematic approach for the discovery of new prions that expands the spectrum of their biological functions.

Biophysical and Mechanistic Models for Disease-Causing Protein Variants.

The rapid decrease in DNA sequencing cost is revolutionizing medicine and science. In medicine, genome sequencing has revealed millions of missense variants that change protein sequences, yet we only understand the molecular and phenotypic consequences of a small fraction. Within protein science, high-throughput deep mutational scanning experiments enable us to probe thousands of variants in a single, multiplexed experiment. We review efforts that bring together these topics via experimental and computational approaches to determine the consequences of missense variants in proteins. We focus on the role of changes in protein stability as a driver for disease, and how experiments, biophysical models, and computation are providing a framework for understanding and predicting how changes in protein sequence affect cellular protein stability.

CRaTER enrichment for on-target gene editing enables generation of variant libraries in hiPSCs.

Standard transgenic cell line generation requires screening 100-1000s of colonies to isolate correctly edited cells. We describe CRISPRa On-Target Editing Retrieval (CRaTER) which enriches for cells with on-target knock-in of a cDNA-fluorescent reporter transgene by transient activation of the targeted locus followed by flow sorting to recover edited cells. We show CRaTER recovers rare cells with heterozygous, biallelic-editing of the transcriptionally-inactive MYH7 locus in human induced pluripotent stem cells (hiPSCs), enriching on average 25-fold compared to standard antibiotic selection. We leveraged CRaTER to enrich for heterozygous knock-in of a library of variants in MYH7, a gene in which missense mutations cause cardiomyopathies, and recovered hiPSCs with 113 different variants. We differentiated these hiPSCs to cardiomyocytes and show MHC-ß fusion proteins can localize as expected. Additionally, single-cell contractility analyses revealed cardiomyocytes with a pathogenic, hypertrophic cardiomyopathy-associated MYH7 variant exhibit salient HCM physiology relative to isogenic controls. Thus, CRaTER substantially reduces screening required for isolation of gene-edited cells, enabling generation of functional transgenic cell lines at unprecedented scale.

Massively parallel variant characterization identifies NUDT15 alleles associated with thiopurine toxicity.

As a prototype of genomics-guided precision medicine, individualized thiopurine dosing based on pharmacogenetics is a highly effective way to mitigate hematopoietic toxicity of this class of drugs. Recently, NUDT15 deficiency was identified as a genetic cause of thiopurine toxicity, and NUDT15-informed preemptive dose reduction was quickly adopted in clinical settings. To exhaustively identify pharmacogenetic variants in this gene, we developed massively parallel NUDT15 function assays to determine the variants' effect on protein abundance and thiopurine cytotoxicity. Of the 3,097 possible missense variants, we characterized the abundance of 2,922 variants and found 54 hotspot residues at which variants resulted in complete loss of protein stability. Analyzing 2,935 variants in the thiopurine cytotoxicity-based assay, we identified 17 additional residues where variants altered NUDT15 activity without affecting protein stability. We identified structural elements key to NUDT15 stability and/or catalytical activity with single amino acid resolution. Functional effects for NUDT15 variants accurately predicted toxicity risk alleles in patients treated with thiopurines with far superior sensitivity and specificity compared to bioinformatic prediction algorithms. In conclusion, our massively parallel variant function assays identified 1,152 deleterious NUDT15 variants, providing a comprehensive reference of variant function and vastly improving the ability to implement pharmacogenetics-guided thiopurine treatment individualization.

Understanding the Origins of Loss of Protein Function by Analyzing the Effects of Thousands of Variants on Activity and Abundance.

Understanding and predicting how amino acid substitutions affect proteins are keys to our basic understanding of protein function and evolution. Amino acid changes may affect protein function in a number of ways including direct perturbations of activity or indirect effects on protein folding and stability. We have analyzed 6,749 experimentally determined variant effects from multiplexed assays on abundance and activity in two proteins (NUDT15 and PTEN) to quantify these effects and find that a third of the variants cause loss of function, and about half of loss-of-function variants also have low cellular abundance. We analyze the structural and mechanistic origins of loss of function and use the experimental data to find residues important for enzymatic activity. We performed computational analyses of protein stability and evolutionary conservation and show how we may predict positions where variants cause loss of activity or abundance. In this way, our results link thermodynamic stability and evolutionary conservation to experimental studies of different properties of protein fitness landscapes.

MaveRegistry: a collaboration platform for multiplexed assays of variant effect.

SUMMARY: Multiplexed assays of variant effect (MAVEs) are capable of experimentally testing all possible single nucleotide or amino acid variants in selected genomic regions, generating 'variant effect maps', which provide biochemical insight and functional evidence to enable more rapid and accurate clinical interpretation of human variation. Because the international community applying MAVE approaches is growing rapidly, we developed the online MaveRegistry platform to catalyze collaboration, reduce redundant efforts, allow stakeholders to nominate targets and enable tracking and sharing of progress on ongoing MAVE projects. AVAILABILITY AND IMPLEMENTATION: MaveRegistry service: https://registry.varianteffect.org. MaveRegistry source code: https://github.com/kvnkuang/maveregistry-front-end.

An improved platform for functional assessment of large protein libraries in mammalian cells.

Multiplex genetic assays can simultaneously test thousands of genetic variants for a property of interest. However, limitations of existing multiplex assay methods in cultured mammalian cells hinder the breadth, speed and scale of these experiments. Here, we describe a series of improvements that greatly enhance the capabilities of a Bxb1 recombinase-based landing pad system for conducting different types of multiplex genetic assays in various mammalian cell lines. We incorporate the landing pad into a lentiviral vector, easing the process of generating new landing pad cell lines. We also develop several new landing pad versions, including one where the Bxb1 recombinase is expressed from the landing pad itself, improving recombination efficiency more than 2-fold and permitting rapid prototyping of transgenic constructs. Other versions incorporate positive and negative selection markers that enable drug-based enrichment of recombinant cells, enabling the use of larger libraries and reducing costs. A version with dual convergent promoters allows enrichment of recombinant cells independent of transgene expression, permitting the assessment of libraries of transgenes that perturb cell growth and survival. Lastly, we demonstrate these improvements by assessing the effects of a combinatorial library of oncogenes and tumor suppressors on cell growth. Collectively, these advancements make multiplex genetic assays in diverse cultured cell lines easier, cheaper and more effective, facilitating future studies probing how proteins impact cell function, using transgenic variant libraries tested individually or in combination.

High-throughput, microscope-based sorting to dissect cellular heterogeneity.

Microscopy is a powerful tool for characterizing complex cellular phenotypes, but linking these phenotypes to genotype or RNA expression at scale remains challenging. Here, we present Visual Cell Sorting, a method that physically separates hundreds of thousands of live cells based on their visual phenotype. Automated imaging and phenotypic analysis directs selective illumination of Dendra2, a photoconvertible fluorescent protein expressed in live cells; these photoactivated cells are then isolated using fluorescence-activated cell sorting. First, we use Visual Cell Sorting to assess hundreds of nuclear localization sequence variants in a pooled format, identifying variants that improve nuclear localization and enabling annotation of nuclear localization sequences in thousands of human proteins. Second, we recover cells that retain normal nuclear morphologies after paclitaxel treatment, and then derive their single-cell transcriptomes to identify pathways associated with paclitaxel resistance in cancers. Unlike alternative methods, Visual Cell Sorting depends on inexpensive reagents and commercially available hardware. As such, it can be readily deployed to uncover the relationships between visual cellular phenotypes and internal states, including genotypes and gene expression programs.

Variant Interpretation: Functional Assays to the Rescue.

Classical genetic approaches for interpreting variants, such as case-control or co-segregation studies, require finding many individuals with each variant. Because the overwhelming majority of variants are present in only a few living humans, this strategy has clear limits. Fully realizing the clinical potential of genetics requires that we accurately infer pathogenicity even for rare or private variation. Many computational approaches to predicting variant effects have been developed, but they can identify only a small fraction of pathogenic variants with the high confidence that is required in the clinic. Experimentally measuring a variant's functional consequences can provide clearer guidance, but individual assays performed only after the discovery of the variant are both time and resource intensive. Here, we discuss how multiplex assays of variant effect (MAVEs) can be used to measure the functional consequences of all possible variants in disease-relevant loci for a variety of molecular and cellular phenotypes. The resulting large-scale functional data can be combined with machine learning and clinical knowledge for the development of "lookup tables" of accurate pathogenicity predictions. A coordinated effort to produce, analyze, and disseminate large-scale functional data generated by multiplex assays could be essential to addressing the variant-interpretation crisis.