Genetics and Genomics 2018

As the costs of genomic analysis decline it becomes possible to offer widescale population genomic analysis to screen for genetic disorders.  The Alabama Genomic Health Initiative was established in 2016 with funding from the State of Alabama to establish such a program on a research basis.  The ultimate goal is to offer genotyping to 10,000 adult volunteers in the state.  After obtaining informed consent, a blood sample is obtained, DNA isolated, and genotyping done using the Illumina Global Screening Array.  Pathogenic or likely pathogenic variants in genes associated with actionable disorders are Sanger verified on a separate blood sample, focusing on the ACMG secondary findings list (currently 59 genes).  Participants are informed of no actionable variants found by mail, or contacted by phone by a genetic counselor if an actionable pathogenic/likely pathogenic variant is identified.  They are then offered further evaluation and management in an appropriate specialty clinic, as well as genetic counseling and facilitation of testing other family members.  Participants are also asked to consent to having their genomic data be maintained in a research database, along with medical information based on their electronic health record and/or questionnaires they complete.  We currently do not return pharmacogenomic data, carrier status, or ancestry-related information, though we are considering doing some of these in the future.  The talk will review progress in the Initiative to date and considerations on future return of results and research questions.

To provide insight on the utility and practicality of implementing a pharmacist-driven pharmacogenomics program in a community pharmacy or primary care setting and to provide a clinical update on common drug-gene implications using clinical case vignettes.

Cell line screening studies require highly efficient protocols for studying many samples in parallel.  We have developed a lyse-and-go protocol for digital gene expression profiling of up to 96-samples by 96-genes in parallel. This protocol is based on nCounter PlexSet Reagents which enable multiplexing of samples and gene targets.  

The PlexSet “lyse-n-go” protocol is simple and fast and can be divided into 3 steps:  1) lyse cells, 2) set-up hybridization with crude lysate; 3) initiate automated post-hybridization processing and digital counting. It does not require RNA purification or cDNA conversion steps, thus saving time and resources for the researcher. 
The data presented are crude cells lysates from various primary cell types, two suspension cell lines, and two adherent cell lines using the IO Lymphocyte Activity Panel to measure the gene expression of 5 housekeeper genes and 91 lymphocyte activity genes (96 genes in total). The data demonstrates that up to 96 high-quality, digital data points for gene expression in each of 96 samples can be generated from crude cell lysates in an efficient, high-throughput protocol and that the cell lysate data correlates well with purified total RNA. 

Structural variants (SVs), defined as rearrangements of genomic sequences, are both a major source of genetic diversity in human populations and are also directly responsible for the pathogenesis of numerous diseases. Many studies have been conducted in the past decade to discover and analyze SVs, however these have predominantly focused on analyzing short-read sequencing data to infer the presence and structure of these variants. A primary limitation shared by all these approaches is their inability to access more complex regions of the genome, particularly repetitive sequences, which preclude the accurately alignment from the shorter read lengths. Recent studies using long-read sequencing technologies have suggested that the genomics community is thus routinely missing tens of thousands of SVs. Here, I outline several efforts to leverage these longer sequences to identify and assess structural variants in previously interrogated well-characterized genomes. I will first describe a new method, PALMER, that uses a pre-masking strategy to identify nested retrotransposition events that have inserted into existing repetitive sequences.  Our early results suggest that as many as 40% of mobile element insertions fall in such regions, and thus their discovery will impact ongoing estimates of mobile element insertion rates. Next, I will present a strategy we recently developed for the high throughput assessment and validation of SVs using recurrence analysis of long-reads. This has enabled the proper interpretation of many false positive or mis-annotated variants called by SV detection algorithms, particularly around complex and repetitive sequences, as well as providing additional support for previously tenuous predictions. We believe these tools will aid the genomics community into better deciphering chromosomal structural rearrangements and furthering our understanding of their mechanistic origins and functional impact.

An ideal microbiome profiling method would produce strain level taxonomic classification of organisms and their relative quantities. We demonstrate proof of principle experiments that sequencing methods using amplicons derived from rRNA gene clusters can be designed to differentiate between known strains of Clostridium difficile, a common hospital acquired pathogen. Obtaining consistent and reproducible results depends on overcoming challenges in microbial lysis, amplicon design, PCR primer degeneracy, sample multiplexing technology, and sequence database analysis. Novel approaches developed for each of these challenges are discussed.

Short-read sequencing has found wide-spread use in metagenomics but the short read lengths and bias with regard to sequence composition have translated to limitations of the study of complex communities. In my talk I will highlight recent advances and applications of long-read PacBio sequencing in this field, both for targeted sequencing as well as shotgun metagenomics, metatranscriptomics and metaepigenomics.

Triple-negative breast cancer (TNBC) has poor prognosis with frequent relapses and deaths using current standard of care treatments.  Metabolic reprograming is now recognized as a fundamental driver of cancer and a detailed understanding of the metabolic rewiring that occurs in TNBC will undoubtedly reveal novel target opportunities. Research will be presented that seeks to use a multi-omics discovery strategy to identify metabolic compensatory in TNBC cells to anti-folate agents.  The overarching goal of our studies is to recognize synthetic lethalities that can be targeted for in TNBC. 

Notably, cell proliferation is critically dependent on tetrahydrofolate (THF) as a critical enzyme cofactor for 1-carbon (1-C) transfer reactions - essential for the synthesis of purines, thymidine and methyl transfer reactions, including DNA, RNA, proteins, lipids and small molecules. Given this essential role of folate, inhibitors of mammalian folate transformation reactions (e.g., methotrexate, 5-fluorouracil, pemetrexed) are widely used for cancer chemotherapy, including TNBC.  Knowledge of compensatory mechanisms to anti-folate therapy could reveal synthetic lethalities, identifying effective new targets in TNBC.  

It is well-appreciated that the major source of 1-C units for support of THF-dependent 1-C reactions is formate, predominantly produced in mitochondria from serine by the sequential enzymatic actions of SHMT2, MTHFD2 and MTHFD1L, followed by formate export to cytosol for support of extra-mitochondrial 1-C biosynthetic reactions. Importantly, mitochondrial synthesis and release of formate relies on a discrete pool of THF in the mitochondria that enters via a selective folate transporter, SLC25A32, and becomes trapped by intra-mitochondrial polyglutamylation. The mitochondrial serine/formate release pathway has received much recent interest as a potential target for cancer chemotherapy, although poorly-defined metabolic compensation has emerged as a concern.  Mitochondrial folate transport by SLC25A32 is yet unstudied for its contribution to oncogenesis, despite knowledge that the slc25a32 gene is amplified in human cancers (including >40% of breast cancers) and this amplification is associated with accelerated disease progression and death. 

Preliminary studies applying a multi-omic discovery approach reveal that slc25a32 gene deletion in TNBC results in a marked cell cycle delay, accompanied by profound metabolic rewiring and perturbed cell signaling pathways. Using multi-omics, we identify compensatory changes in TNBC cells that result from slc25a32 gene knockout, as novel targets for the development of personalized chemotherapy in mitochondrial folate-high cases of TNBC. 

For Research Use Only. Not for use in diagnostic procedures.
 

The All of Us Research Program is a historic effort to gather data from one million or more people living in the United States to accelerate research and improve health. By taking into account individual differences in lifestyle, environment, and biology, researchers will uncover paths toward delivering precision medicine to all Americans. Today, we will discuss the program to date, considerations taken into account during its inception, lessons learned during the beta testing phase, and its trajectory into the future.

We introduce a differential abundance analysis method for the analysis of sparse high-throughput data from large-scale surveys of marker genes for microbial communities. Our approach relies on cumulative sum scaling (CSS) normalization - a count data normalization technique - and the zero-inflated Gaussian (ZIG) model as a statistical method for detecting differential abundance of taxonomic features. ZIG differential abundance detection method accounts for bias introduced by the under-sampling of microbial communities commonly found in large-scale marker gene studies.  We have implemented these methods in the publicly available metagenomeSeq bioconductor package. In addition we highlight the utility of the method in a large scale study characterizing the diarrheal microbiome in young children from developing children. Diarrhea, a major cause of mortality and morbidity in young children from developing countries, leading to as many as 15% of all deaths in children under 5 years of age. While many causes of this disease are already known, conventional diagnostic approaches fail to detect a pathogen in up to 60% of diarrheal cases. Using our novel methodology Streptococci were found in our study to be statistically associated with diarrheal disease in general and more severe forms (such as dysentery) in particular.

Our understanding of how genotype controls phenotype is limited by the scale at which we can precisely alter the genome and assess the phenotypic consequences of each perturbation. In this presentation I will highlight a CRISPR/Cas9-based method in S. cerevisiae for multiplexed accurate genome editing with short, trackable, integrated cellular barcodes (MAGESTIC). MAGESTIC uses array-synthesized oligonucleotides encoding guide RNA-donor DNA pairs with a sophisticated cloning strategy for plasmid-based high-throughput editing. By linearizing the guide-donor plasmid in vivo concomitant with integration at a genomic barcode locus, MAGESTIC circumvents problems associated with post-editing plasmid barcode loss and enables robust phenotyping with one-to-one barcode-to-cell correspondence. We demonstrate that editing efficiency can be increased >5-fold by actively recruiting donor DNA directly to the site of breaks using the LexA-Fkh1p fusion protein. As a proof of principle, we performed saturation editing of the essential gene SEC14 and identified amino acids critical for chemical inhibition of lipid signaling. We also constructed thousands of natural genetic variants, characterized guide mismatch tolerance at the genome-scale, and ascertained that cryptic Pol III termination elements substantially reduce guide efficacy in yeast. MAGESTIC will create opportunities to unravel the genetic basis of quantitative traits, map functional residues on proteins and RNAs across entire pathways, dissect DNA regulatory elements, and build improved organisms for biotechnology.

Tumors are often categorized into standard molecular subtypes. However, largescale studies have demonstrated that patient heterogeneity in the regulatory make-up of tumors remain. At the transcriptional level, one example of heterogeneity in a patient population is the presence of bimodally-expressed genes. Bimodality in expression signifies the presence of potentially new patient sub-groups. Here, we present a new statistical approach called oncomix, that models transcriptional heterogeneity in tumor and adjacent normal (i.e. tumor-free) using bimodality to find oncogene candidates. Oncomix was applied to RNA-sequencing data from the breast cancer cohort of the Cancer Genome Atlas, and a set of oncogene candidates that were over-expressed in only a subset of tumors was identified. 

Intronic DNA methylation was strongly associated with the overexpression of chromobox 2 (CBX2), an oncogene candidate that was identified using our method but not through other approaches. CBX2 overexpression in breast tumors was associated with the upregulation of genes involved in cell cycle progression and is associated with poorer 5-year survival. The predicted function of CBX2 was confirmed in vitro providing the first experimental evidence that CBX2 promotes breast cancer cell growth. Modeling mRNA expression heterogeneity in tumors through bimodal profiles is a novel powerful approach with the potential to uncover therapeutic targets that benefit subsets of cancer patients.

The Knowledge Portal Network is a set of interconnected, open-access web resources that aggregate, harmonize, analyze, and present genetic associations and epigenomic annotations for complex diseases, in order to help researchers elucidate the molecular details of disease mechanisms and find new therapeutic targets. It currently includes the Type 2 Diabetes (type2diabetesgenetics.org), Cardiovascular (broadcvdi.org), and Cerebrovascular (cerebrovascularportal.org) Knowledge Portals. The Portals present an analytical and computational framework whereby researchers can instantaneously search a comprehensive, up-to-date collection of association results and epigenomic annotations relevant to a particular complex disease, all seamlessly integrated with one another, and perform custom association analyses that query securely protected individual-level data. 

In this presentation I will describe results from a family study designed to identify the genetic cause for familial clustering of several early onset cancers.   Data from next generation sequencing combined with expression studies were integrated along with the familial cooccurrence of cancers to identify a genomic region of interest and then to further identify a genetic factor influencing cancer risk.  This factor also interacted with a second factor.  The presentation describes integration of family studies, genomic analysis and expression studies for gene discovery and characterization.  

Complex disease phenotypes - obesity, type II diabetes, and cancer challenge simple models in both evolution and biology.   Examination of molecular networks and their dynamic behavior offers a means to capture gene-centric concepts and complex, emergent behavior.    Examination of patterns of gene variation which appear chaotic and noisy when examined at the individual gene level show coherence when examined in biological network context.  More specifically, in liver cancer, networks provide much stronger signals of disease susceptibility than individual variants. These analyses show emergent, higher-level association of non-syntenic gene variation. When examining multiple obesity and type II diabetes data sets, consistent, recurrent collections of networks predict susceptibility where previous single gene analysis found no overlap.  More provocatively, common susceptibility pathways underpin obesity, type II diabetes, and liver cancer giving clues into disease progression. These interactions demonstrate the emergence of the underlying complex processes important in determining phenotype.  Evolutionary concepts and their related modeling promise to provide critical insight into biologic systems and disease.  In liver cancer, these concepts help explain paths of progression and suggest a novel candidate intervention.

Mesenchymal stem cell (MSC)-derived exosomes mediate tissue regeneration in a variety of diseases including ischemic heart injury, liver fibrosis, and cerebrovascular disease. Despite an increasing number of studies reporting the therapeutic effects of MSC exosomes, the underlying molecular mechanisms and their miRNA complement are poorly characterized. At a system level, miRNA-targeted genes were enriched for (cardio)vascular and angiogenesis processes in line with observed cardiovascular regenerative effects. Targeted pathways were related to Wnt signaling, pro-fibrotic signaling via TGF-ß and PDGF, proliferation, and apoptosis. The data presented in the webinar has helped define the miRNA landscape of MSC exosomes, establish their biological functions through network analyses at a system level, and provide a platform for modulating the overall phenotypic effects of exosomes using nCounter miRNA assays from NanoString Technologies.

Dr. Sigrid Reinsch will discuss NASA’s GeneLab Project whose goal is to develop a unique publicly accessible repository and collaborative workspace that hosts multi-omics datasets generated through biological experiments flown in space. Dr. Reinsch will discuss the mission of GeneLab within the context of NASA’s Space Biology Project and the longer term goals for space exploration. She will then detail how the database is structured, and the unique challenges of generating and analyzing spaceflight omics data. She will discuss the Systems Biology approaches currently used to understand the Omics results from space flight model systems (rodent, fruit fly etc) in the context of tissue response (i.e. muscle, liver, other vital organs). Strategies regarding how GeneLab envisions the involvement of the scientific community and the public at large will also be presented, with current and future capabilities of the system.  Anyone interested in how omics technologies can enable space exploration, or who is curious to understand how space modifies living organisms should attend.

In this presentation I will describe our latest work to obtain comprehensive genomes leveraging long and linked reads. The vast majority of NGS whole-genome data covers hundreds of thousands of samples with short illumina reads, which are unable to capture the full spectrum of genetic variation and genomic complexity. Such comprehensive variation is critical to understanding the full heritability and genetic foundations of human disease. In this seminar I will present our novel alignment strategy (NGMLR) for long read data (Oxford Nanopore and PacBio) and our novel Structural Variations (SVs) caller Sniffles. These two methods improved the accuracy for both technologies enabling the accurate and easy detection of SVs. This includes also nested events that we have previously been blind to or linked events connecting genes over multiple regions.  We will discuss problems, characteristics and limitations of short reads. In addition, I will discuss the impact of these novel found SVs in cancer and other genomes with respect to RNA seq. In the end, I will highlight our current findings where we combine these long read technologies with linked reads to be able to phase SNV and SVs together to obtain a diploid genome per sample. These phased genomes are the most comprehensive representation of genomes up to date and we can now finally generate them within days. 

The arrangement of chromatin inside the nucleus has long been recognized as a key element in genomic stability and an active participant in transcriptional control. With the development and diverse adaptations of the chromosome conformation capture technique (3C), it has become possible to discover many organizational units of the chromatin fiber at the sub-chromosomal and chromosomal scale. Among the most important of these structures are the topologically associating domains (TADs), regions of ~1Mb in size that partition chromosomes into modular structures which are remarkably evolutionarily conserved. Besides their DNA packaging contribution, TADs are recognized as safekeepers of transcriptional control by limiting enhancer-promoter interactions within a defined region, and avoiding awry activation or repression of genes outside the domains.

In this exciting era of “Next-Gen Cytogenetics”, integration of genomic sequencing into the clinical diagnostic setting can provide precise delineation of chromosomal structural rearrangements at nucleotide level. Given the increased risk for congenital abnormalities with de novo balanced chromosome rearrangements, comprehensive interpretation of breakpoints may substantially improve prediction of phenotypic outcomes. A systematic approach for evaluating sequencing results of chromosome rearrangements can be undertaken in light of regulatory chromatin domains: in addition to genes directly located at the breakpoints, dysregulated protein coding genes and non-coding regions can be assessed in relation to TADs, given their role in pathologic rewiring of the human genome through structural rearrangements. Convergent genomic evidence and predicted probability of exhibiting haploinsufficiency is analyzed through publicly available databases including the Database of Chromosomal Imbalance and Phenotype in Humans Using Ensembl Resources (DECIPHER) and Developmental Disorders Genotype-Phenotype Database (DDG2P). Our analyses highlight the important interplay between chromosome organization and disease, and further demonstrate the feasibility of utilizing topological information to predict pathogenic gene-dosage effects. Such analyses can further aid in clinical diagnosis of "n-of-one" samples of non-coding chromosome rearrangements, complementing and enhancing interpretation of current sequencing and array results.

The Genome in a Bottle Consortium (GIAB) has characterized an Ashkenazi trio from the Personal Genome Project (NIST Reference Material 8392) with 12 short, long, and linked read sequencing and mapping methods.  Datasets are public without embargo for analysis and methods development by the community. We have characterized ~3.7 million small variants as well as reference calls for ~90% of the genome with estimated errors of ~2 FPs and 2 FNs per million variants. To extend this characterization to larger indels and structural variants, we collected analyses of variants >=20 bp from 33 bioinformatics methods and five technologies. Nineteen discovery and refinement methods produced sequence-resolved calls using local or global assembly or split reads, giving a precise prediction of deletion breakpoints, inserted sequences, and complex changes.  We designed an integration approach to address challenges in comparing and evaluating large variants, which are frequently in tandem repeats (>50% of all calls) and not precisely characterized. We have iteratively refined our integration process based on feedback received for publicly released draft “straw man” callsets.   Current work includes developing 2-tiered high-confidence variant calls and a high-confidence bed file for benchmarking SVs, as well as a web app for crowd-sourced manual curation of SVs.  These results represent a significant step in GIAB work towards improved benchmarking of large variants in research and clinical settings.

Genomic information has the potential to improve management or potentially prevent or enable early intervention of virtually every disease. However, despite massive technological leaps and exponential decreases in costs, genomics currently plays a relatively niche role in healthcare, with clinical use largely limited to diagnosis of pediatric disorders or as a last resort in guiding cancer treatment. The Garvan Institute’s Kinghorn Centre for Clinical Genomics was one of the first sites in the world to acquire technology to sequence whole human genomes at scale. The service, now incorporated into spinout Genome.One, routinely sequences over 1,000 human genomes per month and is clinically-accredited for diagnostic testing of or predisposition to genetic disease. In addition to serving as a highly effective approach for diagnosing diseases that can be caused by large numbers of different genes, whole genome sequencing has potential for reanalysis in different contexts. This potential argues for a new testing paradigm, where genomic sequencing is undertaken once in an individual’s lifetime and analysed throughout their lifetime to guide clinical decision-making and optimise health management. Here I will present on our implementation, clinical accreditation and performance of whole genome sequencing in the routine diagnosis of genetic and rare diseases and its utility in personal health management and disease prevention. I will also discuss the challenges and opportunities for using genomic information to inform health at point-of-care, the incorporation of genomic information into medical records and its potential to transform medical research.