Ryan D. Hernandez - US grants

DESCRIPTION (provided by applicant): Next Generation Sequencing (NGS) has fundamentally changed the way we study living systems; however, many important questions remain intractable to this powerful approach, due to limitations in read length and accuracy. For example, 95% of all genes in the human transcriptome are thought to be alternatively spliced, with an average of 7 splice junctions per gene. The short reads obtained with current sequencers, however, are unable to span multiple junctions and thus cannot fully characterize this variation. In the study of bacterial and viral evolution and in cataloging tumor-specific chromosomal rearrangements in cancer, accurate long reads are also a key enabling technology. For important biological questions like these to be addressed, new methods are needed that provide more accurate, longer read-length sequencing. The objective of our research is to develop a breakthrough technology to significantly enhance the accuracy and read length of NGS platforms using single-molecule barcoding implemented in droplet-based microfluidics. Each multi-kilobase long molecule will be isolated in a droplet microreactor, amplified, fragmented, and barcoded with a sequence unique to the drop and, thus, to the molecule. Using droplet-based microfluidics, we will barcode thousands of molecules per second-the rate at which these techniques can form, split, inject, and incubate drops. This will allow us to barcode millions of molecules in minutes, far exceeding what is possible with other microfluidic systems and the scale needed to utilize the full capacity of NGS platforms and maximally exploit the barcoding concept. Aim 1: Develop microfluidic hardware to isolate, amplify, fragment, and barcode DNA for sequencing Aim 2: Develop bioinformatics software for DNA reconstruction; validate the approach Impact: Our technology converts the excess depth of short-read deep sequencing into highly accurate long reads. This core capability will have numerous impacts: 1) It will greatly simplify genome assembly by increasing accuracy and read length, allowing currently inaccessible portions of the genome to be sequenced. 2) It will allow a greater fraction of reads to be mapped to scaffolds, reducing the depth of sequencing required to obtain a sequence of a desired coverage, thereby reducing the cost of sequencing. 3) It will allow complete interrogation of splice variation in transcriptomes at the isoform level y allowing transcripts to be sequenced in their entirety with multifold coverage, irrespective of splice structure. 4) It will allow high-confidence identification of chromosomal rearrangements in cancer by increasing sequence accuracy and read length enabling accurate de novo assembly. 5) It will allow investigation of bacterial and viral hyper- evolution in persons with persistent infection by allowing hot spot regions to be sequenced for each microbe individually. Thus, our work will have impacts in genomics, systems biology, cancer, and microbial evolution. Indeed, since our technology markedly increases the read length and accuracy of sequencing platforms, and since sequencing has already had a transformative impact on the biological sciences, we anticipate broad and sustained impacts in basic and clinical areas of research.

DESCRIPTION (provided by applicant): Somatic gene mutations in cancers occur randomly but a significant unresolved question is how selective mutations are retained. Retention is in part determined by fitness traits such as overcoming barriers to proliferation or immune surveillance that confer an adaptation to the tumor microenvironment. Our proposal tests a new idea on fitness traits and adaptation contributing to retention of gene mutations in cancers. This new idea is based on the conserved trait of most cancers having a higher intracellular pH (pHi 7.5-7.6) than normal cells (pHi 7.1-7.2). We predict that selective somatic mutations confer gain or loss of pH sensing that provides an adaptive advantage to the higher pHi of cancers. To limit the scope and risk of testing this new idea we will focus on Arg>His mutations in cancers with the prediction that these mutations encode a gain of pH sensing. Our specific hypothesis is that increased pHi in cancers confers an adaptive advantage for retention of selective arginine to histidine mutations. Our preliminary analysis reveals a higher incidence of Arg>His mutations than expected from codon bias, including CpG mutational effects. Our prediction on gain of pH sensing by Arg>His mutations is based on histidine having a pKa near neutral in solution and hence can be protonated at the pHi of normal cells and uncharged at the higher pHi of cancer cells. In contrast, because Arg has a pKa ~12 it likely remains charged at the pHi of normal and cancer cells. In Aim 1 we test evolutionary forces for retention of histidine mutations by using bioinformatics approaches. We will develop novel bioinformatics techniques for ranking Arg>His somatic mutations, and characterize the historical and contemporaneous evolutionary forces operating at these positions to quantify mutational biases and determine the extent to which pKa altering His mutations are retained in cancers relative to what would be expected across evolutionary timescales. In Aim 2 we experimentally test recurring Arg>His candidate mutations for gain of pH sensing. We will determine pH-dependent functions of wild type and mutant proteins in vitro and in normal and transformed clonal cells. We also will computationally predict pKa's of mutated histidines and arginines and test pH-dependent conformational changes in protein structure using molecular dynamics simulations. To our knowledge mutations in cancers conferring pH sensing, such as p53-R273H, one of the most commonly occurring mutations we will test, has not been reported, nor has the retention of somatic mutations relative to the higher pHi of cancers. If correct, our hypothesis would generate a substantial new view on why some somatic mutations in cancers are retained. Additionally, successfully confirming our hypothesis would be compelling for future studies to test His>Arg mutations for loss of pH-sensing, and mutations involving tyrosine, serine and threonine residues that have pKa's near neutral when phosphorylated. Hence, our findings could be applied broadly and have substantial impact on therapeutic strategies to target pH dependence of mutations driving cancers.

DESCRIPTION (provided by applicant): Genome-wide association studies have been incredibly successful at identifying novel genes and pathways associated with a wide array of complex diseases. However, despite the formation of large consortia to perform meta-analyses across cohorts, only a small fraction of the expected heritability of most common, complex diseases has been explained. The human genetics community is now adopting large-scale sequencing approaches (e.g., exome and whole genome) to identify rare variants that potentially have larger phenotypic effects. In response, statistical geneticists have created a litany of tests for geared toward associating rare variants with disease. We hypothesize that the most parsimonious explanation for an inverse relationship between the frequency of causal alleles and their effect size is that many diseases are caused by an influx of newly arising deleterious mutations that are continually removed from the population due to natural selection. We therefore propose to develop simulation software that will integrate what we know about how allele frequencies change over time from the theory-rich field of population genetics into the data-rich field of human genetics. Our resulting software will be used to develop strategies for sequencing global cohorts with high discovery power, and to aid in the evaluation of existing/future statistical tests. To achieve broad impact, we will create a graphical user interface (GUI) that produces effective figures, and apply our tool to compare and contrast a wide variety of existing statistical tests. We will then revamp our population genetic simulator to become the first population genetic simulator based on the heterogeneous computing architecture of both CPUs and graphical processing units (GPUs). Through intensive parallelization, our software will achieve disruptive efficiency. Using this approach, we will develop a platform for simulation-based inference that can accommodate complex evolutionary models. We will apply this approach to analyze forthcoming whole genome sequencing data from humans and Drosophila. Finally, we aim to return cutting-edge research to the classroom by developing simulation-based teaching tools. Our teaching tool will be in the form of a GUI that enables hands-on learning of complex concepts.

DESCRIPTION (provided by applicant): This competing IMSD grant renewal from the University of California, San Francisco (UCSF) is designed to continue to provide support for students from underrepresented groups to complete the PhD in biomedical and behavioral research and to advance to competitive postdoctoral and academic positions. The proposed program will increase opportunities for underrepresented students to thrive in the biomedical and behavioral sciences by providing them with the necessary tools, skills, and resources to complete their PhDs at UCSF in a timely fashion and to prepare for productive careers in research. Strong academic and professional mentoring leading to the successful completion of PhDs will increase the proportion of underrepresented students who join the biomedical workforce. Robust implementation and evaluation of measurable objectives, milestones, and outcomes will help create a more diverse population of future research scientists. The overall goal of this program will be achieved by the following specific objectives: 1) To ensure that IMSD Fellows' first-year research rotations and thesis lab placements are with renowned faculty in prestigious research programs who are also proven mentors supportive of new students and sensitive to unique issues facing many minority students. Strong mentorship by established researchers is key to the retention of students in the graduate program as well as for meaningful guidance toward obtaining postdoctoral positions and academic appointments after graduation. 2) To promote sustained success of IMSD students throughout their PhD work by fostering a community network of Fellows. Group learning, social engagement, peer mentoring, group discussions with role models, and participation in community service projects will create a support network to promote the progression of the group as a whole. 3) To accelerate IMSD Fellows' academic achievement by developing strong communication and organizational skills in order to maximize their research career progress and success. Progress can be measured by timely completion of milestones toward the degree; success can be measured by Fellows' presentations of their research in local and national settings, awards of competitive extramural fellowships, and eventual publications in high-impact journals. 4) To facilitate Fellows' ability t take a pro-active approach to career planning, by providing them with opportunities for career exploration and with tools and strategies to be competitive in the academic research job market. 5) To nurture a mindset that will encourage IMSD Fellows to serve as mentors, led by example, and support IMSD and similar programs throughout their academic careers by encouraging Fellows who have moved off IMSD funding to remain engaged with the Program through group activities, mentoring of junior fellows, and participation in campus outreach and diversity efforts With multi-faceted institutional support systems at UCSF, this proposed program will help to foster and sustain diversity in tomorrow's biomedical workforce.

DESCRIPTION (provided by applicant): Genome-wide association studies have been incredibly successful at identifying novel genes and pathways associated with a wide array of complex diseases. However, despite the formation of large consortia to perform meta-analyses across cohorts, only a small fraction of the expected heritability of most common, complex diseases has been explained. The human genetics community is now adopting large-scale sequencing approaches (e.g., exome and whole genome) to identify rare variants that potentially have larger phenotypic effects. In response, statistical geneticists have created a litany of tests for geared toward associating rare variants with disease. We hypothesize that the most parsimonious explanation for an inverse relationship between the frequency of causal alleles and their effect size is that many diseases are caused by an influx of newly arising deleterious mutations that are continually removed from the population due to natural selection. We therefore propose to develop simulation software that will integrate what we know about how allele frequencies change over time from the theory-rich field of population genetics into the data-rich field of human genetics. Our resulting software will be used to develop strategies for sequencing global cohorts with high discovery power, and to aid in the evaluation of existing/future statistical tests. To achieve broad impact, we will create a graphical user interface (GUI) that produces effective figures, and apply our tool to compare and contrast a wide variety of existing statistical tests. We will then revamp our population genetic simulator to become the first population genetic simulator based on the heterogeneous computing architecture of both CPUs and graphical processing units (GPUs). Through intensive parallelization, our software will achieve disruptive efficiency. Using this approach, we will develop a platform for simulation-based inference that can accommodate complex evolutionary models. We will apply this approach to analyze forthcoming whole genome sequencing data from humans and Drosophila. Finally, we aim to return cutting-edge research to the classroom by developing simulation-based teaching tools. Our teaching tool will be in the form of a GUI that enables hands-on learning of complex concepts.

ABSTRACT The UCSF Graduate Program in Bioinformatics is seeking continuation of its training program in Bioinformatics and Computational Biology. The program focuses on training young scientists who will serve as leaders in re- search at the interface between computation and biology. The training plan for coursework, enrichment activi- ties, and research reflects the fundamentally collaborative culture at UCSF. Thus, both the formal and informal features of the program have been designed to bring together students from different disciplines and train them for team-based problem solving. Although the focus is on computational research, all students are exposed to experimental biology in many aspects of their training including research rotations and research-based coursework. As a result, our graduates understand the sources of their data as well as how to manipulate it and are prepared to interact in multidisciplinary teams that require an understanding of both wet and dry scientific cultures. Our program has dramatically expanded our outreach efforts to recruit a diverse and talent- ed group of students with computational and quantitative backgrounds, and have developed initiatives to foster a healthy community of inclusion and respect for students across all races, religions, countries of origin, sexual orientations, genders, and ethnicities. We stand with all students, and train them to tackle challenging prob- lems in biology at scales that span the molecular to the phenotypic. The hallmarks of our program include: ? Collaborative and inter-disciplinary research. Training faculty are heavily involved in collaborative research, both within and outside of UCSF. Many of their labs (and trainees) are involved in Consortia or Centers created to address problems that cannot be solved from a single viewpoint but require contributions from many disciplines. Student publications reflect this culture. ? An innovative and evolving curriculum. Our core values of collaboration and interdisciplinary research are instilled from day one in ?Bootcamp?, and continue in well-tested and new intensive project-based core courses designed to establish a common knowledge and language, and to foster team skills. A modular panel of ?selectives? addresses important knowledge gaps in statistics and computer/data science, inherent in the diverse scientific backgrounds of our students. Current and new `mini courses' facilitate deep explo- ration of research topics in small groups with faculty experts, and allow the curriculum to adjust to current scientific developments, and in response to student and alumni feedback and program assessment. ? Intensive training in communication, and preparation for diverse careers. We emphasize training in key competencies needed in diverse careers in academia, industry, or the public sector, including oral and writ- ten presentation, communication, and teamwork skills. Students can participate in career preparation work- shops and internships, and many take on leadership roles in outreach and teaching. Our alumni include leaders in both academia and industry, including several who have started successful companies.