Research Report 2019_20

This HudsonAlpha Institute for Biotechnology research report is a review of the discoveries at the HudsonAlpha labs.

2019-20 Research Report 1 2019-2020 Research Report

601 Genome Way Huntsville AL 35806

To get all the HudsonAlpha research news, visit: www.hudsonalpha.org/research ©2021 All rights reserved.

Introduction ............................................................................ 4

Foundational Genomics . ....................................................... 6 Greg Barsh, MD, PhD and lab. ............................................................. 8 Thomas May, PhD and lab ............................................................. 10 Human Health Research ..................................................... 12 Richard Myers, PhD and lab............................................................... 14 Sara Cooper, PhD and lab. ................................................................. 16 Devin Absher, PhD and lab................................................................. 18 Greg Cooper, PhD and lab.................................................................. 20 David Bick, MD and The Smith Family Clinic for Genomic Medicine... 22 Elaine Lyon, PhD and the HudsonAlpha Clinical Services Lab. ....... 24 Agriscience Research . ........................................................ 26 Jane Grimwood, PhD, Jeremy Schmutz and the HGSC. .................... 28 Kankshita Swaminathan, PhD and lab............................................... 30 Josh Clevenger, PhD and lab. ............................................................ 34 Alex Harkess, PhD and lab ............................................................ 36 Supporting the Institute ...................................................... 38 The HudsonAlpha Health Alliance .................................................. 40 Jian Han, MD, PhD and public and private collaboration ................ 42 COVID-19 and the HudsonAlpha campus innovators response . ...... 44 Neil Lamb, PhD and Educational Outreach .................................... 46 Shawn Levy, PhD and HudsonAlpha Discovery................................... 52

HudsonAlpha Faculty .......................................................... 54

References . .......................................................................... 63

Adjunct Faculty and Scientific Advisory Board . ............... 64

HudsonAlpha 2019-2020 in Review ................................... 66

Discovery, education, and

Thirty years ago, the Human Genome Project marked a new era in science exemplified by the power of collaboration for a common goal, and the importance of communication and open sharing of data. The Hudson- Alpha Institute for Biotechnology was founded on those principles, which continue to resonate broadly across all areas of biology and medicine. We are proud to share with you the accomplishments, progress, and goals of the Institute in this 2019-2020 Research Report. In the two years since our last report, HudsonAlpha scientists published more than 250 papers, recruited two new faculty as part of our Center for Plant Science and Sustainable Agriculture, and continued to make internationally recognized contributions across a diversity of fields, from human health and disease to solving biological mysteries to renewable energy development.

Rick Myers, PhD and colleagues at a Human Genome Project scientific meeting.

a better life for our planet...

This research report also includes our progress and plans for the HudsonAlpha Health Alliance, work carried out by academic partnerships with some of our associate companies, and the tremendous accomplish- ments of the Educational Outreach team. In response to the ever-changing educational landscape brought on by the COVID-19 pandemic, our Educational Outreach team helped meet the needs of virtual classrooms and touched nearly two million individuals through learn- ing experiences and special programs. Together, these areas exemplify the unique environment at Hudson- Alpha, which underlies our vision to leverage the synergy between discovery, education, medicine, and economic development to improve the human condition. As part of our broader response to the global crisis of the COVID-19 pandemic, the Institute has devoted considerable effort to develop new approaches to diagnosis, treatment, and surveillance. Several of our research laboratories and more than ten associate companies that call HudsonAlpha home have made important contributions, and many of these are also included in the report. These institute-wide efforts illuminate the strength of the culture of collaboration at HudsonAlpha, where more than 1,000 scientists, educators, and businesspeople work in 16 non-profit laboratories and 46 associate companies. Our core principles of collaboration towards a common goal and open communication contribute to the unique and rewarding environment of HudsonAlpha. We hope this report captures our passion and commitment to make discoveries, improve human health, and contrib- ute to the future of our community and our planet. n

Greg Barsh, MD, PhD

Greg Barsh, MD, PhD HudsonAlpha Faculty Chair Smith Family Chair in Genomics HudsonAlpha Institute for Biotechnology

Richard M. Myers, PhD President and Science Director M. A. Loya Chair in Genomics HudsonAlpha Institute for Biotechnology

2019-20 Research Report 5

Foundational Genomics

Foundational research, also called basic research, is actually anything but basic. The curiosity and creativity of scientists doing basic research leads to the knowledge base upon which applied research is built. Foundational research expands the horizons in the life and health sciences, and improves the under- standing of biological phenomena. At HudsonAlpha, foundational research includes studies of natural variation, principles of bioethics and computational biology. The knowledge gained from this research is fundamental to our goal of moving advancements in genomics to improve the human condition.

6

HudsonAlpha Institute for Biotechnology

Season 1 of HudsonAlpha's Tiny Expeditions Podcast explores animal morphology, getting into the genetics that make animals look the way they do. From black wolves to striped house cats, HudsonAlpha’s expert on morphology, Greg Barsh, MD, PhD, explains what we know and what we still want to learn.

www.hudsonalpha.org/tinyexpeditions

2019-20 Research Report 7

Greg Barsh, MD, PhD / Barsh Lab

Morphological variation in mammalian color patterns

Zebras, tigers, and cheetahs are all instantly recogniz- able because of their fur patterns and markings. These distinctive patterns serve to camouflage the animals from predators and to help them identify others belong- ing to the same species. The study of the color, structure and form of living animals, called morphology, is a fun- damental aspect of biology. Morphology provides a basis for the understanding of function, development, heredi- ty, taxonomy, ecology and other branches of biology. HudsonAlpha Faculty Investigator Greg Barsh, MD, PhD, is an expert in the genetics of morphological varia- tion, both within and between species. Barsh uses color and color pattern as an experimental platform to study cellular and molecular pathways that are used through- out the body. One goal of the Barsh lab is to understand the development and evolution of periodic color pattern in mammals, to better understand the molecular toolkit that biology uses to generate form and function. Domestic cats are a useful model to study peri- odic color patterns, especially tabby cats whose hair color pattern and form varieties are similar to wild cat species. Barsh and his group previously showed that a gene called Endothelin 3 is expressed at the base of hair follicles in tabby cat markings and plays a key role in the development of tabby pattern 1 . However, tabby markings are apparent in developing hair follicles, indicating that establishment of color pattern must occur at or before hair follicle development. In a new study available on bioRxiv , the group determined when, where, and how, cat color patterns are established during fetal development 2 . Histochem- ical analysis of fetal cat skin revealed stripe-like alter- ations in skin thickness early in fetal development. The skin thickness resembled tabby fur patterns in adult animals. This finding suggests that even before mela- nin-producing cells, called melanocytes, enter the skin, the cells are predestined to signal for a specific fur color. By using single-cell gene expression analysis on fetal cat skin cells, they determined that skin cell expression of a gene called Dickkopf 4 ( Dkk4 ) marks areas of fetal skin that give rise to hair follicles that later

produce dark pigment. Dkk4 -expressing skin cells acquire time-sensitive epigenetic changes that are lat- er incorporated into hair follicles, and ultimately deter- mine whether the underlying skin cells release mole- cules that darken or lighten the hair. The team also showed that two variants in Dkk4 are linked to another genetic locus involved in color pattern- ing, called Ticked . Ticked prevents dark tabby markings, producing hair banding patterns across the entire body surface. Taken together, the results presented in this study confirm a direct role for Dkk4 in cat color pattern establishment, providing a new target for periodic color variation in other mammals. Cats are not the only domesticated animal species with a range of diverse looking members—dogs also have a wide variety of fur colors and color patterns. Barsh is also interested in discovering the genetic basis of coat color in dogs. In a preprint publication avail- able on bioRxiv , Barsh and his colleagues explored the genetic control of pigmentation in dogs, specifically why so many domesticated dogs are yellow, and what genet- ically differentiates a yellow dog from a black dog 3 .

Cat skin histology showing black and yellow hair follicles

8

HudsonAlpha Institute for Biotechnology

FOUNDATIONAL RESEARCH

Modular promoter expression explains Agouti patterns

dominant yellow ASIP allele looks almost the same as the ASIP allele of Arctic grey wolves. Construction of a phylogenetic tree showed that the dominant yellow allele was likely introduced after the grey wolf and arctic grey wolf families split on the tree. Interestingly, they did not find the ASIP hair cycle promoter that controls dominant yellow color in any other living dog. The team concluded that the promot- er was from a ghost lineage, meaning it came from an extinct dog species. Thus, arctic wolves and domestic dogs preserve the genetic legacy of an ice age canid in their pale coats, through natural and artificial selection. The work on cat color patterns was carried out by HudsonAlpha senior scientists Chris Kaelin, PhD and Kelly McGowan, PhD; the work on dog and wolf an- cestry was carried out by Chris Kaelin, PhD in collab- oration with the research groups of Tosso Leeb, PhD (University of Bern) and Danika Bannasch, PhD, DVM (UC Davis). Taken together, this work provides new in- sight into biologic building blocks that give rise to period- ic body structures such as digits and vertebrae (as well as stripes and spots), and helps us understand the origin and evolution of the natural world in which we live. n

In mammals, specific color patterns arise through differential regulation of a gene called Agouti ( ASIP ) which encodes a signaling molecule that causes hair follicle melanocytes to switch from making eumelanin (black or brown) to pheomelanin (yellow to nearly white). Through analysis of skin RNA-seq data from dogs with different coat patterns, Barsh and his colleagues deter- mined that variants in two promoters of ASIP , ventral promoter (VP) and hair color promoter (HCP), control coat color in different parts of the dog’s body. Genetic variation in the promoters explains five dis- tinctive dog color patterns—dominant yellow, shaded yellow, agouti, black saddle, and black back. For exam- ple, dogs with black back coloration have solid black bodies but yellow paws and faces. Lack of expression of any one of several variants of hair cycle promoter turns off ASIP expression in the body of the dogs, while ex- pression of ventral promoters turns on ASIP expression in the paws and face. Barsh’s lab expanded this analysis to include modern and ancient wild dogs and uncovered the evolutionary history of the promoter. Through ge- nome sequence analysis, they determined that the

2019-20 Research Report 9

Thomas May, PhD / Bioethics

Increasing diversity among genomic research participants and databases

Conducting genomic research in diverse populations facilitates the understanding of health disparities, new discoveries in biology, more accurate matching of diverse patients with safe and effective treatments, improved interpretation of genetic tests, and better tracing of human history. Despite these clear benefits, individuals from diverse ethnic backgrounds have been widely underrepresented, or even absent, in genomics research. Failure to include sufficient diversity among research participants denies some populations full participation in the benefits of research findings, like pharmacogenomics, and inhibits richer understanding of gene-disease relationships among all populations. Despite efforts to increase diversity in genomic research, only marginal progress has been made toward meeting the goal. The proportion of participants in genome-wide association studies (GWAS) of non- European ancestry increased from four to nineteen percent between 2009 and 2016, but almost all of this gain in diversity was attributed to participants of Asian ancestry. In the United States, lack of representation of African Americans among genomic research par- ticipants is of particular concern. Persons of African descent make up 12 to 14 percent of the U.S. population, yet they continue to make up fewer than four percent of GWAS participants.

HudsonAlpha Faculty Investigator Thomas May, PhD, is a bioethicist who is dedicated to increas- ing diversity among genomic research participants by identifying and addressing the reasons minori- ties lack representation and participation in genom- ic research. In an opinion piece published in Trends in Genetics in 2020, May discusses the numerous reasons why minority groups may choose not to participate in genomics research 1 . May points to two main factors that deter minori- ties from participating in genomics research: lack of trust in research purposes, and the inadequacy of privacy protection. Many minorities are skeptical to give consent to al- low their DNA to be released to large research databas- es because they fear the data could be used in studies that go against their beliefs or morals. The distrust in research purposes is justified based upon historical events like the Tuskegee Syphilis Study in which African Americans were denied treatment for syphilis in order to advance a research study, and the Havasupai Tribe lawsuit in which DNA samples from tribe members were used for genetics research outside of what they had initially consented. The story of HeLa cell research and the inability of the family of Henrietta Lacks (from whom these cells were obtained) to benefit due to lack of insurance coverage also compounds the distrust in biological and genomic research. May states that given the history of unethical medical research on minority communities, researchers should consider more lenient opt-out procedures for genomic databases for studies where minority participants are targeted. Researchers should also work to increase a sense of control and en- gender trust with the minority populations. Privacy concerns surrounding genomic research are also of important significance to minority groups. The re- cent rise in the use of genetic databanks as a forensic tool for law enforcement raises questions about the potential for this tool to oppress minority groups. This is enough to provide disincentives for some minority populations. There is also a concern that data gleaned from genomic

2009

2016

373 studies 1.7 million samples

2,511 studies 35 million samples

81% European ancestry

96% European ancestry

Asian

Other, Non-Asian

4%

19%

Non-European ancestry

Reference: Popejoy, A.B. and Fullerton, S.M. Genomics is failing on diversity. Nature . 538:161-164 (2016)

10

HudsonAlpha Institute for Biotechnology

FOUNDATIONAL RESEARCH

Total AGHI recruitment by the numbers

The Alabama Genomic Health Initiative (AGHI) is a partnership between HudsonAlpha and UAB-Medicine to provide genomic testing, interpretation, and counseling free of charge to residents in each of Alabama’s 67 counties. The AGHI also includes a major focus on research, through which data from test results will be used to advance scien- tific understanding of the role that genes play in health and disease. Visit www.uabmedicine.org/aghi to learn more.

Enrollment by race:

White....................................................................... 71.2% Black or African American...................................... 20.7% Hispanic/Latino....................................................... 3.72% More Than One Race. ................................................ 3.4% Asian. ........................................................................ 2.3% Unknown................................................................... 2.1% Native American Indian or Alaska Native.................. 0.3% Native Hawaiian or Other Pacific Islander. ............... 0.4% 72% are FEMALE and 28% MALE

To date, 5,369 individuals, representing all 67 Alabama counties, have been enrolled in the population screening group of AGHI. 81 positive genotyping results among

In order to recruit a participant cohort reflective of Alabama's rich diversity, the recruitment team de- veloped community engagement strategies that took into consideration the fundamental challenges to recruitment of African American participants. By host- ing health fairs and recruitment events at churches and community organizations throughout the state, the AGHI team recruited diverse populations from every county in Alabama. While African American participation has not reached the representation of this community as a percentage of Alabama's overall population, AGHI has achieved an overall representation exceeding 20 percent for African Americans to date. The AGHI team hopes that their published engage- ment techniques and recruitment strategies, which resulted in significant improvement in representation of African American participants in Alabama, can be adopted in other states. By continuing to gain the trust of minority populations in genomic research, the research- ers hope that genomics research will one day include di- verse, multi-ethnic populations to accurately represent genetics-related disease risks in all populations. n 80 individuals (1.5 percent) were identified in the population cohort. These results include risk-increasing variants for hereditary cancer, cardiomyopathy, malignant hyperthermia and hypercholesterolemia.

research could discriminate individuals in employment and insurability. In his opinion piece, May concludes that addressing the underlying distrust in genomics research and privacy policies is an important step in increasing the diversity in genomics research participation. May and his colleagues set out to achieve greater diversity in genomic research participation and address the fundamental challenges to recruiting minority pop- ulations through the Alabama Genomic Health Initiative (AGHI) 2 . With an African American population constitut- ing about 26 to 27 percent of the state’s overall popula- tion, Alabama is well positioned to help meet this goal.

2019-20 Research Report 11

Human Health Research

Understanding the human genome and how it is involved in health and disease are important goals for HudsonAlpha scientists and clinicians. Whether they are identifying gene variants that cause rare childhood disorders, applying new technology to find proteins involved in chemotherapy resistance, or understanding how genome modifications contribute to autoimmune conditions, HudsonAlpha research- ers are passionate about harnessing the power of genomics and genetics to attack human disease. Collaboration and data-sharing are also hallmarks of our research because they help to advance our mission to improve human lives and well-being.

12

HudsonAlpha Institute for Biotechnology

2019-20 Research Report 13

Richard M. Myers, PhD / Myers Lab

Gene expression and the human brain

identity and precise location of many functional elements in the human genome, such as non-protein encoding genes, promoters, and transcriptional regula- tory sequences, remain to be fully elucidated. The Ency- clopedia of DNA Elements (ENCODE) Project began af- ter the Human Genome Project with the goal of building a comprehensive list of functional elements in the hu- man genome. Myers’ lab has been major contributors to the ENCODE Project. At HudsonAlpha, his group collab- orates with HudsonAlpha Adjunct Faculty member Eric Mendenhall, PhD and his lab to generate large amounts of data that help explain how human genes are regulated. For the third phase of the ENCODE Project, Myers’ lab studied transcription factor biology in genome-wide experiments to expand the knowledge of DNA elements in both human and mouse genomes 1 . The project, over- seen by project manager Mark Mackiewicz, PhD and Chris Partridge, PhD, studied about a quarter of all of the expressed transcription factors in a human liver cancer cell line. It was the largest study of transcription factors expressed at physiological, or normal, levels to date. The researchers identified novel associations be- tween transcription factors, elaborated on their spatial interactions on DNA, and distinguished between those that interact with promoters and those that interact with enhancers in the genome. With the help of Mendenhall’s lab, this study continues to grow, more than doubling the number of transcription factors and adding additional functional data.

Although all of an individual’s cells contain the same DNA, each cell can become specific for a certain tis- sue or organ system by turning the expression of dif- ferent genes on and off. Gene expression is regulated by the binding of transcription factor proteins to short stretches of DNA, called regulatory elements, that serve as on/off switches for that gene. The regulation of gene expression controls the timing, location, and amount of gene products in a cell. This allows for the differen- tiation and development of unique cell types through- out the body, but can also lead to disease states if it becomes dysregulated.

Richard M. Myers, PhD in the lab

HudsonAlpha Faculty Investigator Richard M. Myers, PhD, and his lab study the human genome with the goal of understanding how changes in gene expres- sion contribute to human health and disease, as well as to basic biological processes. Myers’ group develops and applies innovative technologies and high-through- put next-generation sequencing technology to identify, characterize and understand gene regulatory systems. By first understanding gene expression and regulatory systems in a healthy state, his group then studies how they are altered during the development and progres- sion of human disease. Although a near-complete human genome se- quence was finished in 2003, with contributions from the genome center Myers directed at Stanford, the

Mark Mackiewicz, PhD studying transcription factors

14

HudsonAlpha Institute for Biotechnology

HUMAN HEALTH

Brian Roberts and Chris Partridge, PhD

Jacob Loupe, PhD

Lindsay RIzzardi, PhD

Matthew Knuesel, PhD and Sophie Guo

Bri Rodgers and Nick Cochran, PhD

A few of the members of the Myers Lab working on human health research

The goal after identifying transcription factors is to use a variety of approaches to turn down the expression of mutated genes in hopes of mitigating the effects of the disease. Several HudsonAlpha labs, including Myers’ lab, along with several other institutes are sequencing and analyzing thousands of patient samples from with- in the state of Alabama and around the world to learn more about the genetic causes of diseases of the ner- vous system. These studies have helped identify genes involved in ALS 3 , Alzheimer disease 4 , FTD and bipolar disorder 5 . In one such study, Cochran helped collabora- tors in California and Colombia, South America find a never-before-identified mutation on PSEN1 , a gene known to cause Alzheimer Disease 4 . This finding is helping improve efficacy in clinical trials for drugs to treat the disease. By gaining a better understanding of the genetic and genomic bases of neurological disorders, Myers’ lab hopes to identify genes and transcriptional pathways that will lead to predictive biomarkers for the disorders and responses to treatments, as well as new targets for therapies. A project led by postdoctoral fellow Ben Henderson is searching for RNA and DNA markers in plasma from individuals with Alzheimer disease and Huntington disease that are indicative of the presence of the disease, even prior to onset of symptoms. n

In a related study published in Genome Research , Myers’ lab, along with Sara Cooper’s lab at Hudson- Alpha, analyzed ENCODE data to characterize regions of DNA with high numbers of transcription factor binding events, called high occupancy target (HOT) loci 2 . Clas- sically, gene regulatory loci are thought to be bound by only a small subset of transcription factors. However, by using two methods to find DNA binding associations across the genome, the group found about 15,000 HOT loci that could be bound by more than twenty five per- cent of the DNA binding proteins assayed. The study also presents evidence that regulatory activity is localized to a few hundred base pairs within each HOT locus. Myers’ lab also studies particular transcription factors and networks of factors involved in human brain biology and disorders of the nervous system, such as Huntington disease, Alzheimer disease, fron- tal temporal dementia (FTD), ALS, bipolar disor- der, schizophrenia, and ALS. Members of Myers’ lab, including senior scientists Brian Roberts, Nick Cochran, PhD, Jacob Loupe, PhD, Lindsay Rizzardi, PhD, Matthew Knuesel, PhD, and graduate stu- dent Bri Rogers, are using a variety of experimental approaches and cultured human neuronal cells to iden- tify transcription factors and their binding sites in and near genes responsible for neurodegenerative diseases, particularly Huntington disease and Alzheimer disease.

2019-20 Research Report 15

Sara Cooper, PhD / Cooper Lab

Predicting multi-drug resistance in cancer using CRISPR technology

Although many cancers are initially susceptible to chemotherapy, over time they can develop resistance and stop responding to the treatment. This resistance is what leads to progression of the disease and is the ultimate cause of death for many cancer patients. The development of multiple drug resistance (MDR) is one of the major challenges in cancer treatment and represents the leading cause of treatment failure. Cancer cells acquire MDR in many ways. Scientists have identified changes in cancer cells grown in a dish that change how they respond to drugs. Some of these changes include increased expression of transporter pumps that remove the drugs from the cancer cell, met- abolic changes that break down the drug, and changes in cell cycle checkpoints that would normally prevent cell division in the presence of certain drugs. However, the translation of this knowledge to the clinic has not been widely successful. Therefore, identifying genes and mechanisms critical to the development of MDR and establishing a reliable method for detecting them in clinical samples could help predict the development of resistance and lead to treatments designed to avoid it.

their pancreatic cancer study were published to the preprint server bioRxiv in October 2020 1 . The group used CRISPR gene editing technology to activate and knockout genes in pancreatic cancer cell lines. They conducted four genome-wide CRISPR activation and knockout screens to identify genes whose loss or gain of expression were able to alter sensitivity to four of the most common chemotherapies used in the treat- ment of pancreatic cancer (gemcitabine, 5-fluorouracil, irinotecan, and oxaliplatin).

CRISPR activation of the cellular transport pump ABCG2 induced resistance in cell lines across each of the drug treatments. ABCG2 has been associated with multiple drug resistance in several previous studies. However, when the group looked at ABCG2 expression in pancreatic tumors from patients it was not high- ly expressed, suggesting it does not have significant relevance in patients. Further analysis of the CRISPR screen identified additional genes that were relevant for resistance in patient tissues. Using an algorithm developed in their lab, the group computed drug sen- sitivity scores based on expression levels of resis- tance genes, and separated cell lines and patients into different treatment response groups. Genes identified by the CRISPR screen could be used to predict drug sensitivity in cell lines and patients based on their gene expression profiles and direct personalized therapeutic approaches.

Sara Cooper, PhD in the lab

At the HudsonAlpha Institute for Biotechnolo- gy, Faculty Investigator Sara Cooper, PhD, and her lab are using genome-wide CRISPR screening methods to identify genes associated with chemotherapy resis- tance in pancreatic and ovarian cancer. Results from

16

HudsonAlpha Institute for Biotechnology

HUMAN HEALTH

Emily Gordon, PhD working on cancer tumor sensitivity

informed treatment decisions. In addition to their work in pancreatic cancer, the lab is also using a sim- ilar CRISPR screening approach to study resistance mechanisms in ovarian cancer. Led by senior scientist Emily Gordon, PhD, the group aims to predict tumor sensitivity to drugs in combinations often seen in the treatment of ovarian cancer by screening 2,400 genes using the CRISPR activation/knock out system. The results of these experiments will be used to interpret gene expression data gathered from patient tumors pre- and post-treatment and better understand and predict treatment response. Overall, the Cooper lab is dedicated to the iden- tification of novel therapeutic targets, and strategies for treating pancreatic and ovarian cancer, both of which have some of the worst outcomes among cancer patients. Embracing the collaborative spirit of HudsonAlpha, Cooper entered into a joint project with resident company CFD Research. The company has the tools and expertise necessary to solve protein structures and predict drugs that might bind them. Using the three-dimensional structure of the protein, CFD Research can predict which existing chemical compounds might be able to attach to it and render it non-functional. In their collaboration, CFD Research has worked to identify novel inhibitors of the gene ANGPTL4 , a target identified by Cooper and Myers (see page 14 for Myers Lab studies) in 2016. If resis- tance-inducing proteins can safely be turned off using a drug, it could increase the efficacy and success of traditional cancer therapies. n

Cooper’s lab also identified gene pathways as- sociated with drug resistance based on their CRISPR screens. Activation of chromatin remodeling path- ways was one of the most consistent mechanisms of drug resistance identified in the screens. Chromatin remodeling is the dynamic modification of tightly wound DNA, called chromatin, from its condensed, or closed, state to an open state that allows access to the machinery necessary to turn genes on. Widespread closed chromatin has been previously associated with poor prognosis in pancreatic cancer patients. Coo- per’s lab explored the role of one protein involved in chromatin remodeling that is also associated with pa- tient prognosis. Histone deacetylase 1 (HDAC1) is an enzyme that cooperates with chromatin repressors. Overexpressing HDAC1 in cell lines regulated genes that are part of the epithelial-to-mesenchymal tran- sition, a pathway known to be involved in multi-drug resistance. These data further the understanding of the role of HDAC1 and chromatin remodeling in drug resistance and provides important information in the pursuit of improving treatment options for pancreatic cancer. PhD candidate Carter Wright is continuing to pursue the additional work aimed at identifying some of the key genes and pathways downstream of HDAC1 activation that might represent novel therapeutic targets themselves. The data generated through the CRISPR screen completed in Cooper’s lab provides a promising new tool to predict a patient’s likelihood of developing re- sistance to common treatments, and help make

2019-20 Research Report 17

Devin Absher, PhD / Absher Lab

The role of epigenetics in lupus frequency and severity

Because females carry two X chromosomes com- pared to men who carry one, mammals evolved X chro- mosome inactivation to randomly silence genes on one of the X chromosomes in females. This allows for equal expression of X-linked genes across sexes. X chromo- some inactivation is largely mediated by the non-coding RNA molecule XIST, which recruits multiple proteins to the inactivated X chromosome to initiate and maintain gene silencing. By using RNA sequencing technology, the team compared expression patterns of X-linked alleles in immune cells from female lupus patients and healthy subjects. They found that SLE patients had bias in expression of X-linked genes in T and B cells, which may indicate a problem with X-inactivation. Offering a potential mechanism for the biased ex- pression of X-linked genes, they also observed that XIST was upregulated in the immune cells of patients with SLE. While additional studies are still needed to prove aberrant X chromosome inactivation, these findings present promising results that may help explain the female predisposition to autoimmune diseases like SLE.

Autoimmune diseases are a complex and mys- terious set of diseases. They present a particu- lar challenge to researchers seeking to define the cause and explain the progression of the diseases. While the study of the human genome, especially genome-wide association studies, has been essential in contributing to the current understanding of autoim- mune diseases, translating that knowledge into mech- anistic insight and disease treatments still remains a challenge. Epigenetics has emerged as an important additional layer of instructions that control how DNA is interpret- ed. Several studies have implicated epigenetic influenc- es in the development of autoimmunity, although there is still much to be elucidated. HudsonAlpha Faculty Investigator Devin Absher, PhD, is a leader in the field of epigenetics. Absher’s work focuses on how changes in the epigenome influ- ence human disease. His lab has previously studied cardiovascular disease and the dietary and metabolic risk factors for heart disease, cancer epigenetics, and the effects of aging on the epigenome. Currently, Absh- er is focused on understanding the role of epigenetics in autoimmune disease, specifically systemic lupus erythematosus (SLE). SLE is a complex autoimmune disease that is char- acterized by a dysregulated immune system, leading to widespread, chronic inflammation and tissue damage. It commonly affects the joints, skin, brain, lungs, kidneys, and blood vessels. Because the underlying mechanisms driving the disease are poorly understood, there is no cure and treatments focus simply on alleviat- ing symptoms of the disease. Like many other autoimmune diseases, SLE dispro- portionately affects women more than men. Although it affects women nearly 10 times more frequently than men, the molecular basis for the sex disparity is widely unknown. In a recent study published in Human Molecular Genetics , Absher and his team present strong evidence that X chromosome inactivation could be involved in the disparity 1 .

Kevin Roberts, PhD preparing systems that help study SLE

18

HudsonAlpha Institute for Biotechnology

HUMAN HEALTH

Devin Absher and Jun Song of the Absher Lab are focused on understanding the role of epigenetics in autoimmune disease, specifically systemic lupus erythematosus (SLE) They found that epigenetic defects in female African American patients with SLE were already present in immature B cells emerging from the bone mar- row, whereas epigenetic defects in female European American patients with SLE appeared later in B cell development. Furthermore, they showed that CpGs near IFN- regulated genes are hypomethylated in African Amer- ican patients with SLE from the earliest circulating B cell stage, indicating that B cells might be epigeneti- cally “primed” for an aberrant immune response pri- or to maturation in SLE patients of African American ancestry. These epigenetic differences may explain the differences in disease presentation and severity between the ethnic groups as well. These studies add to the breadth of work from Absher’s lab that reinforce the importance of epigenetics in understanding human health and disease. Through the work of Absher and others at the Institute, Hudson- Alpha continues to contribute to the growing body of knowledge about epigenetics. n

In addition to affecting women at a higher rate than men, epidemiologic studies have revealed that SLE also affects African American women with greater frequency and severity than women of European and Asian ethnic- ities. Epigenetic dysregulation of B cell differentiation is thought to be an important mechanism underlying the pathogenesis of SLE, however, the majority of epi- genetic SLE studies to date have focused on European and Asian populations. Absher’s lab sought to remedy this deficiency in diverse data sets by studying African American women with SLE 2 . The team used various statistical and machine- learning methods to better understand the biologic changes that occur throughout B cell development in SLE patients and to discern any differences in these ef- fects between African American and European Ameri- can subjects. They analyzed whole-genome DNA meth- ylation data from B cell subsets in African American and European American females with and without SLE. Results from the study indicate that SLE- specific methylation patterns are ethnicity dependent.

2019-20 Research Report 19

Greg Cooper, PhD / Cooper Lab

Genomic approaches to understanding rare disease

association of genetic variation in BRSK2 with neurode- velopmental disorders 1 . While information gleaned from next-generation sequencing technology has led to many diagnoses, researchers like Cooper are constantly looking for ways to improve the technology and increase diagnostic rates for patients with rare disease. In a recent publication in Human genetics and Genomics Advances 2 , the Cooper lab describes how an exciting new technology, called long- read sequencing, helped them identify pathogenic variants responsible for previously undiagnosable, rare neurodevelopmental disorders. Recent advances in long-read sequencing technolo- gy have allowed for the production of reads, or sequenc- es, that are up to 100 times longer than those resulting from the widely-used short-read sequencing technol- ogy, from Illumina ® , the leading sequencing company. This leads to fewer gaps in the whole sequence once the pieces are assembled. Using long- read sequencing, the team, led by senior scientist Susan Hiatt, PhD, ana- lyzed six family trios (mom, dad, and the affected child) that each had children affected by neurodevelopmental disorders. Although the families had each previously had their genomes sequenced with short-read technol- ogy, no causal genetic variant had been identified. Byusinglong-readsequencing, theyfoundthousands of genetic variants in each family that had previously been missed. Among these newly detected variants, the team identified likely pathogenic variation in two of the six children. This study shows the ability of long-read se- quencing to capture complex variation missed by short- read sequencing. It is likely that it will become a powerful front-line tool for research and clinical testingwithin rare disease genetics. Cooper’s team is also focused on scaling up rare disease diagnosis through the use of genetic screening tests. A recent study led by senior scientist Kevin Bowling, PhD, was published in Genetics in Medicine and highlights findings on the effectiveness of

Two out of every 100 children are born with physical disability or developmental delay, which often arise from genetic factors. However, because many of the underly- ing diseases are so rare, specific diagnoses for children with developmental delay are elusive. Identifying genet- ic variants, whether de novo (not inherited from either parent) or inherited, can provide a disease diagnosis, guide treatment approaches, and give families the an- swer to their years-long medical mystery. By using genome sequencing technology to identify rare variants associated with these diseases, the lab of HudsonAlpha Faculty Investigator Greg Cooper, PhD, is able to help end the diagnostic odyssey for many chil- dren with developmental delay. With collaborators like the University of Alabama at Birmingham, the Clinical Sequencing Evidence-Generating Research (CSER) Consortium and the Alabama Genomic Health Initiative (AGHI), members of Cooper’s lab have sequenced more than 1,467 affected children and approximately 1,511 parents. They have found genetic cause for about twen- ty-seven percent of the affected children, leading to a more precise and definitive clinical diagnosis. In addition to helping physicians diagnose rare disease in children, identifying rare variants is also invaluable to the research community. Members of Cooper’s lab use GeneMatcher , a web service that allows researchers to share genes of interest in order to collaborate with other scientists to solve genetic mysteries. They have submitted 166 genes to GeneMatcher , which has led to 18 collaborative publications since 2016. The Cooper lab discovered variations in one such gene, BRSK2 , in several children with devel- opmental delay or intellectual disability but wanted more instances with which to compare. Using Gen- eMatcher , five more individuals with variations in the gene were identified and compared to one an- other. Through statistical and biological analysis of the mutations in BRSK2 , the group confirmed the

20

HudsonAlpha Institute for Biotechnology

HUMAN HEALTH

L to R Kevin Bowling, PhD, Greg Cooper, PhD, and Susan Hiatt, PhD looking at variations in genes to help diagnose rare disease in children.

disease-related genetic variant when they do not, has potentially damaging repercussions such as emotional distress and unnecessary medical interventions. By analyzing self-reported race and ethnicity data, the group also found that African-American individ- uals were more likely than European Americans to receive a false positive result. This likely reflects reduced representation of non-European individuals in clin- ical and research genetic databases. Together, the observations from this study support the notion that array detected disease-linked variants, such as those detectedingeneticscreeningarrays,shouldbeconfirmed by an additional method in a clinical genetics lab prior to returning results to patients, especially for underrepre- sented minorities. The continued diagnostic success of the Cooper lab coupled with the identification of disease-linked genet- ic variants not only provide clarity and hope to patients and their families, but also add to the growing catalog of knowledge of genetic mechanisms that contribute to human health and disease. n

population-based genomic screening in Alabama 3 . At the time of publication, the team had screened 5,369 Alabamians in an attempt to identify genetic varia- tion that may be relevant to participants’ health. The genetic screens were conducted using a screening array designed to genotype more than 654,000 variants, including ~160,000 rare variants that may increase an individual’s susceptibility to a disease. In order for genetic screening tests to be used to inform people of potential health risk, they must be accurate and reliable. While medically-relevant variants were confirmed in ~1.5 percent of study participants, an additional ~1.5 percent of participants were flagged as having medically-relevant variants that turned out to be false positives, meaning they could not be confirmed by DNA sequencing. If confirmatory testing had not been conducted, roughlyhalf of the individualswitharray-iden- tified medically-relevant findings would have been false- ly told that they had an increased risk of a serious condi- tion like cancer or heart disease. Providing false positive results to an individual, or telling them they have a

2019-20 Research Report 21

David Bick, MD / The Smith Family Clinic for Genomic Medicine

Best practices for clinical whole-genome sequencing

Advances in next-generation sequencing technology are revolutionizing rare genetic disease diagnosis. The most comprehensive sequencing technology, called whole-genome sequencing (WGS), is a powerful clinical tool for detecting known and potential disease-caus- ing DNA variations. However, chromosomal microarray analysis and whole-exome sequencing, not WGS, are currently indicated as first-tier tests for the diagnosis of many rare genetic diseases.

Through his role at The Smith Family Clinic for Genomic Medicine, Bick has witnessed firsthand the value of using WGS to solve cases of misdiagnosed and undiagnosed genetic disease. Patients referred to the clinic have usually been on a diagnostic odyssey— searching for answers for many years, visiting dozens of doctors, and some receiving misdiagnoses. WGS al- lows geneticists like Bick to look at all of the patient’s genes at once, greatly increasing the likelihood of making a diagnosis. However, a lack of standards for best-in-class clin- ical WGS presents a barrier to its widespread adoption. Because of HudsonAlpha’s expertise using WGS as a clinical test, Bick was chosen to serve as a member of the Medical Genome Initiative (MGI), which consists of genomic medicine experts who work tirelessly to over- come this barrier. The MGI aims to expand access to high-quality clinical WGS for the diagnosis of genetic diseases through the publication of common laboratory and clinical best practices. In a paper published in December 2020, Bick and the MGI team presented consensus recommenda- tions and suggested best practices for the analytical validation of clinical WGS for the diagnosis of germline genetic disease 1 . Analytical validation demonstrates the accuracy, precision, and reproducibility of a test, which is abundantly important for a test returning results as complicated as WGS. Based upon their practical experi- ences and discussions amongst the group, they present- ed key statements describing their recommendations for analytical validation of clinical WGS. Their recommendations ranged from the type and number of variants analyzed and reported in a WGS test to the type of reference standards that should be used to validate each run of a WGS test. Importantly, the group endorsed the use of clinical WGS as a viable first-tier test for rare disorders, and suggested that it replace chromosomal microarray analysis and whole-exome sequencing as a first-tier test. Bick and the MGI team hope that the practical advice put forth in the publication will encourage other laboratories to introduce WGS into

Dr. Bick working on interpreting a whole genome sequence on HudsonAlpha Codicem software.

David Bick, MD, believes that clinical WGS is ready to take over as a first-tier test for rare genetic disease diagnosis. Bick is a leader in the genomic medicine field and serves as the chief medical officer and a faculty investigator at the Institute, the medical director of The Smith Family Clinic for Genomic Medicine LLC , and a laboratory director in the HudsonAlpha Clinical Services Laboratory, LLC .

22

HudsonAlpha Institute for Biotechnology

HUMAN HEALTH

The HudsonAlpha Clinical Sevices Lab performs whole genome sequencing for The Smith Family Clinic for Genomic Medicine.

clinical practice. They also acknowledge that best practices will continue to evolve and hope this publica- tion will open more discussions focused on guidelines for analytical validation of clinical WGS. In addition to ensuring that clinical WGS can accu- rately and reproducibly diagnose a given disease, evi- dence of clinical utility and cost-effectiveness will also be important for WGS to be accepted into routine prac- tice. In another paper published in October 2020, Bick and the MGI team provide a standardized framework and toolkit of best practices for measuring the clinical utility of WGS 2 . Although they focused on its application for rare germline disease, the team hopes the clinical utility toolkit will offer a flexible framework for best practices around measuring clinical utility for a range of WGS applications. While progress is being made in defining best prac- tices for the implementation of clinical WGS, the same guidelines cannot be used for other genomic tests, such as elective screening tests. Elective genomic screening tests, meaning a test performed on individuals who do not have a medical indication for such testing, are gain- ing popularity. Using guidelines specific to other genetic tests could lead to a higher risk of false-positive reports because they are performed on a different population with a different intended use. The Smith Family Clinic for Genomic Medicine offers both types of testing, giving Bick the opportunity to document successes and challenges of implement- ing and analyzing elective genomic screening tests

Dr. Bick at The Smith Family Clinic for Genomic Medicine.

as well as WGS. Bick authored a perspective piece, published in The Journal of Molecular Diagnostics , that addresses the application of basic test design principles to genetic screening 3 . In the perspective piece, Bick sug- gested that laboratories can minimize the risk of false- positives by modeling estimates of missing data (like natural history and prevalence of diseases), by design- ing tests according to their intended use, by adhering to established principles of screening, and by providing consumers and physicians abundantly clear limitations to the clinical utility of the results. Under the leadership of Bick, The Smith Family Clinic for Genomic Medicine affords patients access to cutting-edge genomic technology and also helps inform researchers, clinicians, and policy-makers on how that same genomic technology can be implemented to aid patients on their health or diagnostic journey. n

2019-20 Research Report 23

Page 1 Page 2 Page 3 Page 4 Page 5 Page 6 Page 7 Page 8 Page 9 Page 10 Page 11 Page 12 Page 13 Page 14 Page 15 Page 16 Page 17 Page 18 Page 19 Page 20 Page 21 Page 22 Page 23 Page 24 Page 25 Page 26 Page 27 Page 28 Page 29 Page 30 Page 31 Page 32 Page 33 Page 34 Page 35 Page 36 Page 37 Page 38 Page 39 Page 40 Page 41 Page 42 Page 43 Page 44 Page 45 Page 46 Page 47 Page 48 Page 49 Page 50 Page 51 Page 52 Page 53 Page 54 Page 55 Page 56 Page 57 Page 58 Page 59 Page 60 Page 61 Page 62 Page 63 Page 64 Page 65 Page 66 Page 67 Page 68

hudsonalpha.org

Made with FlippingBook - Online magazine maker