2021 CCIB Annual Retreat – Center for Computational and Integrative Biology

Tuesday, December 14, 2021
9:00 a.m. – 8:00 p.m.

Location: Camden County Boathouse
7050 N. Park Drive, Pennsauken, NJ 08109

2021 CCIB RETREAT – ITINERARY

Tuesday, December 14, 2021
9:00am	CCIB Retreat Meet–N–Greet Location: Camden County Boathouse Explore CCIB Retreat and Greet Others.

9:30am	WELCOME Director, CCIB – Dr. Nir Yakoby Rutgers-Camden Provost – Dr. Daniel Hart Interim Dean, Faculty of Arts and Sciences–Camden – Dr. Bob Atkins Associate Dean, Graduate School Faculty of Arts and Sciences–Camden – Dr. Michelle Meloy Senior Vice President for Research, Rutgers University – Dr. Michael Zwick

10:00am	KEYNOTE SPEAKER: DR. SUDHIR KUMAR LH Carnell Professor, Director, iGEM, Department of Biology & Department of Computer & Information Sciences at Temple University TITLE: Evolutionary Technologies: Phylogenomics and Phylomedicine ABSTRACT: New mutations continuously arise in the genomes of cells and organisms. Comparative analysis of sequences from individuals, populations, and species yields patterns of evolutionary conservation and divergence. These patterns have become the primary tool to reconstruct the cellular histories of tumors and pathogens as well as the tree of life. The next generation sequencing has fast-tracked these efforts, putting data-driven discovery and hypothesis testing on steroids. I will discuss our new evolutionary techniques for analyzing very large sequence datasets in phylogenomics (tree of life) to phylomedicine (tumors and coronavirus). These methods have high accuracy, fast computational times, and small memory footprints. They will democratize big data analytics, improve scientific rigor, and make scientific pursuit more broadly accessible.
11:00am	BREAK

11:15pm	BEST STUDENT PAPER PRESENTATION SPEAKER: Heather Ciallella, CCIB Ph.D. Student, Brannigan Lab Title: Predictive modeling of estrogen receptor agonist, antagonism, and binding activities using machine and deep-learning approaches. SPEAKER: Stacy Love, CCIB Ph.D. Student, Salas Lab Title: Facile treatment to fine-tune cellulose crystals in cellulose-silk bio composites through hydrogen peroxide.

12:00pm	LUNCH

1:00pm	STUDENT POSTER SESSION

4:00pm	FACULTY SPOTLIGHT SPEAKER: Dr. Angelica Gonzalez TITLE: Biodiversity Patterns and Processes: Looking to the Past to Understand the Future of Biodiversity SPEAKER: Dr Grace Brannigan TITLE: Contiguously-hydrophobic sequences are frequently significant throughout the human exome.

5:00pm	WORKSHOPS Faculty Breakout Session: The role of CCIB as a Center and future expectations. LOCATION: Masters Room Post Doc Breakout Session: Meet and Greet LOCATION: Lounge Student Breakout Session: Preparing for Industry LOCATION: Main Auditorium

6:00pm 7:00pm	DINNER CCIB SOCIAL

STUDENT POSTER SESSION: TITLES & ABSTRACTS

SID BHADRA-LOBO, LAMOUREUX LAB

TITLE: Dock2D: Toy datasets for the molecular recognition problem

ABSTRACT: Predicting the physical interaction of proteins is a cornerstone problem in computational biology. New classes of learning-based algorithms are actively being developed, and are typically trained end-to-end on protein structures extracted from the Protein Data Bank. These training datasets tend to be large and difficult to use for prototyping and, unlike image or natural language datasets, they are not easily interpretable by non-experts. We propose Dock2D-IP and Dock2D-FI, two toy datasets that can be used to select algorithms predicting protein-protein interactions—or any other type of molecular interactions. Using two-dimensional shapes as input, each example from Dock2D-FI describes the fact of interaction between two shapes and each example from Dock2D-IP describes the interaction pose of two shapes known to interact.With the hope that it will stimulate further research, we also propose a number of baselines that represent different approaches to the problem.

JOE BRAASCH, GONZALEZ LAB

TITLE: Soil chemistry heterogeneity explains genomic variation in a rare, endemic herb

ABSTRACT: Small population sizes and restricted distributions are expected to increase the likelihood of extirpation through evolutionary and ecological drift and contribute to loss of genetic variation. Yet, ‘rarity’, is a common ecological strategy and some of the most diverse plant groups are primarily comprised of rare species. Calochortus tiburonensis is a perennial herb with a global distribution of roughly 0.5km², encompassing multiple patches on a single hill in Marin County, CA. Historical records in this botanical hotspot suggest that this is natural distribution of C. tiburonensis, and not the result of contraction after habitat loss, making it an exceptional system to study the evolutionary ecology of rarity. C. tiburonensis grows on stressful soils which have low quantities of essential nutrients and high heavy metal content. We expected that soil properties could explain the patchy distribution of C. tiburonensis and if so, patches could also exhibit differences in genomic variation. To test this hypothesis, we conducted surveys of soil chemistry and genomic variation in C. tiburonensis. Surprisingly, we found there was no spatial segregation of soil chemistry or genomic variation. Yet, soil chemistry was able to explain 57% of the total genomic variation. Additionally, C. tiburonensis had higher genomic variation than expected, which could be the product of balancing selection imposed through high soil chemistry variation. As a next step, we will test for balancing selection by 1) comparing genetic variation summary statistics to previously published studies that use similar methods, and 2) analysis of the site frequency spectrum. These results suggest that spatial heterogeneity can be an important mechanism in preserving genetic variation in rare species.

SIENNA CASCIATO, LEE LAB

TITLE: Characterizing Genetic Mechanisms for Measuring Day-Length in Neurospora crassa

ABSTRACT: There are predictable changes most of the organisms on Earth experience based on the daily rotation of the Earth and the early rotation of the Earth around the sun. The biological rhythms with a period of about one day is called circadian rhythm and the seasonal rhythm with a period of one year is called circannual rhythm. Because ambient light condition is one of the strongest environmental cues, the predictable seasonal rhythm of a day length must play an important role in developing the circannual rhythm. Photoperiodism is a physiological response of an organism to changes of the ambient environment over a year and plays a major role in fitness of an organism in nature. For the past half century, the molecular mechanisms of the circadian clockhave been characterized. However, the mechanisms of the circannual clock and the role of the circadian clock in the circannual clock have not been understood. We hypothesized that there are multiple genes that are involved in photoperiodism, and that the genes involved in the circadian clock might also be involved in photoperiodism. To test our hypotheses, we developed the protoperithecia assay (PPA). Protoperithecia is a female sexual reproductive structure in N. crassa. It is known that the number of protoperithecia changes in response to different day-length periods. We performed Quantitative Trait Loci (QTL) analysis on the change of the number of photoperiod-dependent protoperithecia as a trait in 91 F1 progeny of N. crassa. We found a major QTL on chromosome (Chr) 1 and characterized 17 knockout mutants in the target region on Chr 1. We identified a candidate QTL gene whose deletion had a significant deviation of the photoperiodic response from that of the parent genotype. We also performed PPA on 10 known classical circadian clock mutants and found that some of the clock genes showed altered photoperiodic responses. Our data support the view that the circadian clock is a part of the day-length measuring mechanism, and thus involved in the circannual clock.

JOSHUA CHAMBERLAIN, KLEIN LAB

TITLE: Bacterial Sphingolipids: the Impact of Sphingolipid on Membrane Formation, Structure, and Stability.

ABSTRACT: While sphingolipids are found ubiquitously in the membranes of eukaryotic cells, they have only been observed in a handful of prokaryotic taxa. Additionally, the functions of sphingolipids in prokaryotes are less well-defined than in eukaryotes. Nevertheless, it has been established that the addition and removal of sphingolipids has a measurable impact on the bacterial membrane. We aim to use the impact of sphingolipids on the membrane to further our abilities to create custom synthetic vesicles. By altering the variety of sphingolipids in a vesicle membrane, we can give the vesicle specific properties, such as resistance to temperature or susceptibility to changes in osmotic pressure. This is accomplished by utilizing the recently discovered sphingolipid synthetic pathway of Caulobacter crescentus. We can induce the production of many different bacterial sphingolipids in C. crescentus, extract the total lipids in bulk, and create vesicles in vitro for characterization. In this study we detail our methods for creating nanovesicles, the contributions of these nanovesicles to our efforts, and their limitations as a model system for sphingolipid characterization.

ELENA CHUNG, ZHU LAB

TITLE: A Data-Driven Computational Modeling Study to Identify and Prioritize Potential Human Carcinogens Using Public Data

ABSTRACT: Currently, tens of thousands of chemicals in daily use lack detailed toxicity information, posing a risk to human health and the environment. Traditional risk assessment methodologies require expensive and time-consuming studies to identify chemicals with complex toxicities, such as carcinogens. Computational approaches, such as quantitative structure-activity relationship (QSAR) models offer a promising alternative strategy for chemical risk assessment. Available QSAR models can predict new compounds for their simple physicochemical properties and simple biological activities. However, QSAR models normally cannot predict new compounds well for complex toxicity endpoints such as carcinogenicity. We developed a data-driven computational approach to develop QSAR models for carcinogenicity-related assays. To this goal, a probe dataset was obtained from the United States Environmental Protection Agency’s Integrated Risk Information System (IRIS), containing the carcinogenicity information of 342 chemicals. This dataset was used to identify relevant bioassay data from PubChem (http://pubchem.ncbi.nlm.nih.gov/). Among the 1,971 PubChem assays with testing results for the chemicals in the IRIS dataset, 25 assays were identified as relevant to carcinogenicity (p < 0.05) and were selected to develop QSAR models. The tested chemicals in each bioassay were used to create 15 QSAR models using the combination of five machine learning algorithms and three chemical descriptors, totaling 375 models across the 25 assays. Among the 25 bioassays, eight resulted in QSAR models with acceptable predictivities based on 5-fold cross-validation (i.e., balanced accuracies of 69.05% ± 5.88%). Using our QSAR models, we can predict and rank the carcinogenic potentials of 342 IRIS compounds correctly (i.e., positive predictive values of 75.99% ± 7.07%). We also used these models to predict five external databases of pesticides, drugs, high production volume chemicals, natural products, and cosmetics. The results showed that more potential carcinogens exist in the pesticides and drug databases than in the other three. This study portends an automated modeling technique that can be applied to prioritize new potential carcinogens by employing validated QSAR models from different assays obtained from a public data resource. Supported by P30ES005022, R15ES023148, R01ES029275, and R01ES031080.

CADEN COMSA, GENEVA LAB

TITLE: Genomic Powerhouse: Phylogenetics of Anole Mitochondria

ABSTRACT: Anole lizards, belonging to the monophyletic genus Anolis, are a model clade for the fields of evolutionary biology, ecology, and their intersection. However, despite extensive study of the genus, several fundamental questions about the diversification and evolution of the over 425 species in the Anolis clade still exist. We set out to investigate the basal relationships between the 8 major subclades of anoles, as well as to test for monophyly of the genus, and non-monophyly of species belonging to the same ecomorphological class (ecomorphs). We created a pipeline to process short-read sequencing files and generate mitochondrial (mito-)genomes. We implemented the pipeline on whole genome sequencing (WGS) reads from our own datasets, as well as on WGS and Target Capture sequencing (TC) data publicly available on the Short Reads Archive (SRA) website from the National Center for Biotechnology Information (NCBI). We then performed phylogenetic analyses with all currently available Anolis mito-genomes. Wesuccessfully de novo assembled 18 complete mito-genomes using our pipeline, including one genome from off-target reads pulled from a TC dataset. We confirmed the monophyly of Anolisand the non-monophyly of the ecomorphs, as well as found support for the monophyly of the 8 subclades. We also found that the structure of the mito-genomes was conserved across all analyzed species. Improving our understanding of the evolutionary history of Anolis will allow for a better understanding of how this incredibly diverse clade evolved and open up new avenues for hypothesis-driven research. This pipeline also provides an easy and accessible way to assemble mito-genomes from large short-read sequencing data files, as well as for repurposing preexisting, publicly available data to further leverage its contribution to science.

ALEXA DERAGO, PICCOLI & GONZALEZ LABS

TITLE: Computational Workflow for Ancient Network Reconstruction

ABSTRACT: A network approach to studying paleoecological data can provide us with information on how past communities responded to change. This information can then be used to make predictions on how present day communities may adapt to changing conditions. Rodent middens, preserved plant matter and arthropod remains, offer a good source for paleoecological data. They can be used as records of past assemblages. The goal of this project is to create a computational framework for the reconstruction of ancient networks using paleomidden data. This framework will take into consideration predator-prey relations using body mass ratios, spatial distance of collected samples, and the time period from which the samples were preserved. The workflow will also be constructed so that it is easily extendable for use with any independently collected paleomidden data.

TRUMAN DUNKLEY, KLEIN LAB

TITLE: Mechanisms Involved in Uropathogenic E. coli Pathogenesis

ABSTRACT: Urinary tract infections are primarily caused by uropathogenic E. coli which invade bladder umbrella epithelial cells, and following endosomal escape, rapidly proliferate within the host cell before effluxing and infecting neighboring cells. While this mechanism has been extensively studied within murine models, attempts to examine this invasion pathway in vitro have proven somewhat unsuccessful due to the bacteria remaining trapped within Lamp1+ endosomes. Previous attempts to facilitate this escape have worked with some success, through pharmacological disruption of the actin cytoskeleton via addition of cytochalasin D. Since extracellular matrix stiffness has been shown to be a modulator of the actin network, we believe that alterations to this stiffness can therefore be used to promote bacterial endosomal escape. Indeed, our data suggests that 5637 bladder epithelial cells, when grown on physiologically relevant polyacrylamide gels and infected with uropathogenic bacteria, have a greater rate of bacterial endosomal escape and proliferation within the host cell.

JAHMAL ENNIS, BRANNIGAN LAB

TITLE: Effects of Lipid Membrane Composition on Gold Nanoparticle Aggregation

ABSTRACT: Liposome nanoparticle hybrid carriers are promising drug delivery systems. We observe aggregation of gold nanoparticles in these hybrid vesicles. Aggregation shifts the resonance frequency of GNP’s, limiting control over vesicle rupture. Therefore, we designed 2 sets of simulations to observe the effect of membrane composition on nanoparticle aggregation. Increased long range ordering is observed with increases in AuNP concentration and in lipids with long acyl chains.

QIN HE, Post DOc, qi laB

TITLE: Auxin receptor gene OsAFB6, a positive regulator of panicle size and grain yield in rice

ABSTRACT: Auxin plays critical roles in many developmental processes of plants. The auxin signaling pathway isa series of plant responses to auxin stimuli. As auxin receptors, TIR/AFB family genes encode F-Box proteins that directly bind auxin and then transduce the stimulus through the signaling pathway. In this paper, we found that overexpression of OsAFB6 delayed heading by upregulating OsRR1and downregulating Ehd1expression levels, increased cytokinin (CK) and simultaneously decreased the IAA concentration in the young panicle, which promoted inflorescence meristem development and resulted in large panicles with more spikelets per panicle, primary branches and increased grain.

LAUREN HINKEL, KLEIN LAB

TITLE: Sphingolipid Metabolism in Gram Negative Bacteria

ABSTRACT: Sphingolipids are amphipathic antimicrobial lipids that have many biological functions. While most extensively studied in eukaryotes, sphingolipids are found in many organisms including bacteria. Sphingolipids are frequently encountered by bacteria as they are present, to varying degrees, in many bacterial niches. Although genes putatively involved in sphingolipid metabolism have been identified in a wide array of bacteria few have been thoroughly characterized. Sphingosine-responsive genes were identified through RNA-seq analysis in the gram-negative bacterium Pseudomonas aeruginosa PAO1. The protein sequences of Pseudomonas sphingosine-responsive genes were used to identify homologs in Caulobacter crescentus NA1000. Homologous recombination was used to create targeted gene deletions and microbial culturing techniques were used to determine phenotypic differences between strains grown in the presence or absence of antimicrobial sphingosine. The putative sphingosine metabolic genes identified in P. aeruginosa, sphBCD, are essential for maximal Pseudomonas growth in the presence of sphingosine. Similarly, the sphC homolog CCNA_01277 of C. crescentus contributes to Caulobacter‘s growth in the presence of sphingosine.

XUELIAN JIA, ZHU LAB

TITLE: Mechanism-Driven Modeling of Drug-Induced Liver Injury Using Structural Alerts and a High Throughput Screening Assay

ABSTRACT: Evaluation of preclinical and clinical DILI is expensive, time-consuming, and often requires many subjects to detect hepatotoxicity signals. In vitro testing assays are not sufficient to predict complex chemical toxicities such as DILI. As an advanced framework of risk assessment, Adverse Outcome Pathways (AOPs) describe mechanisms that involve complex chemical toxicities. We aimed to apply a computational framework to developing an AOP that is predictive of DILI. For this effort, we employed complementary computational modeling and in vitro assays focusing on oxidative stress via activation of the antioxidant response element (ARE). We collected and curated a dataset of drug compounds (n=704) with known DILI labels as modeling set, which were used to search PubChem for relevant assay data. The ARE assay, which identifies chemicals that have the potential to induce oxidative stress, shows high correlations to DILI (PPV=0.82). Next, we developed quantitative structure-activity relationship (QSAR) models based on ARE data to predict the oxidative stress response for compounds lacking ARE test results. Furthermore, potential toxicity alerts were identified from chemical fragments that correlated to DILI. The mechanistic DILI model consists of the identified structural alerts as molecular initiating events and in vitroARE activation as a key event. To experimentally validate the ARE QSAR model and the resulted mechanistic DILI model, 11 compounds in the modeling set and 12 new external compounds were selected and experimentally tested using our in-house ARE screening assay. The mechanistic model showed good DILI predictivity (accuracy = 0.78) for modeling set and new compounds. Potential false positive DILI predictions caused by only using ARE results can be corrected by incorporating the alerts and vice versa. This mechanistic model illustrates a potential toxicity pathway from initial chemical features to associated cellular responses, and this strategy can be applied to develop predictive models for other complex toxicity endpoints.

AMAN KAUR, QI LAB

TITLE: Exploring the mechanism of stomata formation under drought stress

ABSTRACT: Stomata are micropores on the epidermis of the aerial land plants. They open and close to facilitate gas exchange, including CO2 uptake for photosynthesis and water evaporation for transpiration. Under drought conditions, many plants inhibit stromal formation to preserve water, and increase water-use-efficiency. Our understanding on the mechanism of stomatal formation has greatly advanced in the last two decades, thanks to studies in the model plant Arabidopsis thaliana. A typical molecular pathway, including the EPIDERMAL PATTERNING FACTOR (EPF) ligands, the receptor complex composed of the ERECTA family receptor kinases and the TMM receptor protein, a MAPK cascade and several bHLH transcription factors, have been identified to regulate stomatal formation. Using a combination of approaches in genetics and cell biology, we aim to explore the mechanism of drought regulation on the stomatal formation pathway.

JAMES KELLEY, GRIGORIEV LAB

TITLE: Systematic identification of SVs with non-primary alignments

ABSTRACT: WGS data is typically analyzed to find single nucleotide variants (SNVs), indels, structural variants (SVs), and copy number variants (CNVs). Variant callers can produce enormous amounts of data. Analysis can be automated for SNVs and indels; automation is difficult for SVs and CNVs. Numerous algorithms have been developed to characterize CNVs/SVs in NGS data, their poor performance results in reliance on manual curation. Algorithms for detecting SVs include read pair (RP), comparison of insert size to calculated range; read depth (RD); and split reads (SR), only one read in the pair has a reliable mapping. In SR analysis, reads contain soft-clipping, filtered out in many pipelines. However, soft-clipped portions should not always be considered bad data, especially in the context of SR calls. They could generate non-primary, or supplementary alignments, used as evidence of SVs. For example, inversion reads can be represented as a primary alignment with an inverted supplementary alignment. We are utilizing this in the development of GROMSOM somatic caller based on our GROM algorithm (osf.io/6rtws), which comprehensively detects all variant types in a single run with superior speed and accuracy compared to other variant callers. GROMSOM is a basis of our pipeline to analyze complex SVs using visualization to compare RP, SR, and RD data since these algorithms when used together can strengthen the evidence of an SV call. A visual representation of read data allows variant calls to be evaluated by manual inspection, as demonstrated by the popularity of Integrative Genomics Viewer (IGV) software. We are extending the representation of non-primary alignments in our pipeline to improve the identification of complex SVs. We have analyzed hundreds of samples and produced case studies to demonstrate the value of our approach. Our pipeline should be useful in resolving complex genome rearrangements such as those foundin cancer samples.

TESS KONNOVITCH, GONZALEZ LAB

TITLE: The Effects of Warming and Nutrients on Rates of Ecological Change

ABSTRACT: Human activities are significantly affecting the structure and functioning of aquatic and terrestrial ecosystems. Changes in climate and nutrient inputs have strong implications on ecological phenomena through their effects on rates and pathways of energy and matter transfer in ecosystems. Current studies show contrasting effects of nutrient and warming on processes from organisms to ecosystems, and advancing ecological understanding relies on understanding generalities and identifying the mechanisms underlying context-dependent responses. We are currently using a global meta-analysis to test the hypotheses that warming and nutrients increase ecological process rates (e.g., growth, production, consumption), and that the magnitude of these responses are nutrient-dependent. Preliminary results suggest that heterotrophic rates increase with warming and nutrients, whereas autotrophic rates are not significantly altered. Heterotrophs are known to be less flexible in their nutrient requirements, which is one hypothesis as to why they experience a strong rate increase in response to changing nutrient availability. Additionally, heterotrophic processes are more temperature-dependent as they are driven by ATP synthesis, whereas autotrophic processes are driven by Rubisco carboxylation and less temperature-dependent. Moving forward, we plan to determine the role of trophic level (e.g., herbivores and carnivores), habitat (aquatic vs. terrestrial), and chemical element (nitrogen, phosphorus) to understand whether these factors may help explain some heterogeneity in our observed responses.

MARK NESSEL, Gonzalez LAB

TITLE: Decline of insects and arachnids driven by nutrient enrichment: a meta-analysis

ABSTRACT: Recent studies have documented global declines in insects and their relatives, but the exact mechanisms for these declines are not fully understood. Potential drivers underlying these declines are increases in anthropogenic inputs of nitrogen (N) and phosphorus (P). Here, we present a meta-analysis of 901 cases from 84 studies reporting the effects of N, P, and combined N+P enrichment on the abundance of hexapods (insects and collembola) and arachnids. We found that N and combined N+P enrichment caused significant decreases in the abundance of these groups overall. Enrichment effects were observed in aquatic and terrestrial hexapods, and arachnids, in temperate as well as tropical habitats. The abundance of above- and below-ground arthropods decreased significantly after N+P addition, while N addition reduced above-ground but not below-ground arthropods. The negative effects of nutrient enrichment on arthropods were consistent regardless of their metamorphosis type. Our results also show that the negative effects of N+P additions are stronger for aquatic insects associated with ecosystem health (Ephemeroptera, Plecoptera, and Trichoptera) than for other insects. These findings suggest that changes in elemental cycles are a potential cause of the ongoing global decline of insects and other closely related arthropods and underscore the serious effects of nutrient enrichment on ecological systems.

SUNG WON OH, FU LAB

TITLE: dNanosis for Rapid Detection of Nucleic Acid

ABSTRACT: Using the novel DNA-mediated proximity assembly circuit (DPAC) of biochemical reactions, dNanosis (DNA Nano Diagnosis) platform is developed which can detect various bio-targets and report easy-to-read signals. The assembly circuit is regulated by a DNA logic-AND-gate module, which is comprised of a DNA hairpin-locked catalytic cofactor and a toehold. Targets of nucleic acids, small molecules, or proteins trigger the conformational switch of DPAC by dynamic mechanisms of toehold-mediated strand displacement and expose the cofactor. When enzyme/cofactor pair actuates a reaction, colorimetric or fluorescence signals are produced and detected. We are currently developing dNanosis assay platform which integrates RNA extraction, isothermal amplification, and DPAC sensor for rapid detection of current viral COVID-19 virus.

JOHN ORGERA, GRIGORIEV LAB

TITLE: Finding Evidence of Translated Structural Variants in Clinical Samples of SARS-Cov-2

ABSTRACT: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the virus behind the current COVID-19 pandemic. To further understand its replicational ability we have analyzed its production of non-canonical junctions. We used publicly available Mass Spectrometry data of clinical COVID-19 cases to test for the translation of these junctions. Previous studies have not shown their presence in clinical cases, only cell culture. These junctions could create novel proteins or be targets for the host’s defenses.

SOMYA PATRO, LEE LAB

TITLE: Characterizing the Genetic and Molecular Mechanisms of GlcAT-P in Circadian and Photoperiodic Regulation in Drosophila melanogaster

ABSTRACT: A circannual rhythm is a self-sustained endogenous rhythm driving biological oscillations that persist with a regular periodicity of about 12 months, meaning that it has an annual rhythm. Photoperiodism is an example of a circannual rhythm and refers to the response an organism has to seasonal variations in day-length. Seasonal changes have a major impact on most organisms’ growth and development, and their ability to reproduce and survive in nature. Circadian rhythm is a self-sustained endogenous rhythm that regulates the timing of daily events within an organism and has a period of close to 24 hours. It has been previously shown that severe changes in photoperiods and circadian misalignment could negatively impact human health, causing several diseases and disorders. In this study, we will characterize the genetic and molecular mechanisms of the GlcAT-P knockout mutant involved in circadian clock and photoperiodic regulation, using Drosophila melanogaster. We also aim to identify any shared regulators between these two time-measuring phenomena. The overall goal is to characterize genes responsible for measuring photoperiodism and better understand how the Drosophila melanogaster measures changes in day-length as well as how they decide to initiate a photoperiodic response. To conduct this study, we will utilize previously established assays to determine the role of GlcAT-P in photoperiodism. First, we will conduct the chill coma recovery (CCR) assay wherein a comatose state is induced in the flies after keeping them at lethally low temperatures. We will calculate their recovery time from this comatose state and use this data to perform a Cohen’s D analysis to measure the effect size. Next, we will conduct the diapause assay. Diapause refers to a period of suspended development that occurs as a response to seasonal changes. We will examine the level of ovarian development under different photoperiods, using dissection. We will also conduct these assays under Nanda-Hamner conditions of constant light but increasing hours of darkness. Using these experimental methods, we hope to successfully determine the genetic and molecular mechanisms of GlcAT-P involved in photoperiodic regulation in insects, and possibly pave the way for future research regarding photoperiodic regulation affecting human health.

CONNOR PITMAN, BRANNIGAN LAB

TITLE: The Role of Methionine-Methionine Interactions in the Conformations of Intrinsically Disordered Proteins.

ABSTRACT: Intrinsically disordered proteins (IDPs) lack well defined structures. These proteins play many roles within cells, from signaling, to acting as chaperones, to interacting with several binding partners. Because they do not fall into a well-defined conformation, it is a challenge for researchers to understand the effects of mutations on the overall structure of IDPs. In this poster, we investigate the effects of mutations involving methionine on the overall structure of the IDP beta amyloid through the use of molecular dynamics simulations. We find evidence for specific methionine-methionine interactions, and demonstrate how the mutations affect how segments of the protein interact with each other.

ADAM POFF, YAKOBY LAB

TITLE:Is tRNA:Val-CAC-1-1 a tRNA or a tRF?

ABSTRACT: Drosophila melanogaster has seven Val-CAC tRNAs. It was previously demonstrated that the removal of tRNA:Val-CAC-1-1 cannot be tolerated as a homozygous deletion. We aim to understand whether the tRNA:Val-CAC-1-1 is required as a tRNA or for its regulatory role as tRNA fragment (tRF). tRFs are single stranded RNAs produced by specific post transcriptional cleavage of tRNA by Dicer or other nucleases. These fragments then enter the RNAi pathway and function similar to miRNAs. tRNA:Val-CAC-1-1 has been predicted to be a tRF for numerous genes, including Sprouty, an important inhibitor of the epidermal growth factor receptor (EGFR) signaling pathway. Using the genome engineering technique, CRISPR/Cas9, we will mutate the predicted core tRF as well as the anticodon of tRNA:Val-CAC-1-1. Transgenic flies will be tested for their ability to become homozygous. In addition, due to the regulatory role of EGFR in oogenesis, eggshell morphology will be documented. The proposed experiments will discern the function of this gene as tRNA or tRF.

MOTLEY REX, GRIGORIEV LAB

TITLE: Predicting Putative Regulatory Targets of Ribosomal RNA Fragments: a Machine Learning Approach

ABSTRACT: Recent studies have suggested that ribosomal RNA-derived fragments (rRFs) may have a role forming silencing complexes with Argonaute for the purpose of post-transcriptional gene regulation, in a similar fashion to microRNAs (miRNAs). Though targets of regulation by miRNA are well-characterized, few rRFs have been studied at length, and to our knowledge there is not yet a database or predictive tool characterizing rRF targets. Here, we expand on a previous study of putative rRF-target pairs in a Crosslinking, Ligation, and Sequencing of Hybrids (CLASH) dataset of Argonaute 1 pulldowns in HEK293 cells. We train a Gradient Boosting Classifier (GBC) machine learning algorithm on CLASH rRF/transcript pairs using features such as nucleotide makeup, binding energy, and secondary structure, with the intent of predicting other potential targets of rRF-mediated regulation. Such a predictive tool will help to narrow the focus of potential experimental studies to rRFs which are likely to have regulatory effects of particular interest.

EZRY SANTIAGO-McRAE

TITLE: Anchoring Synthetic DNA to a Membrane: A Novel Application of Free Energy Perturbations

SNEHA SEELAM, SALAS LAB

TITLE: Characterization of Keratin-Cellulose Bio-Composites Coagulated in Varying Solutions of Hydrogen Peroxide

ABSTRACT: Biomaterials is a rapidly growing field of study which can be used to replace or substitute our existing materials to achieve a more eco-friendly one. They also have a unique advantage, where they can be conditioned to our benefits by tuning their morphological and physicochemical properties, unlike existing materials. Cellulose is a popular choice within this field due its low cost, abundance, crystalline and amorphous properties, whereas keratin is made up of fibrous protein, secondary structures, found mainly in our hair, skin and nails and distinct properties like biodegradability, biocompatibility and mechanical durability. Hydrogen Peroxide is widely used as a coagulation agent during fabrication method as it’s known to affect the crystallinity of the bio-composites, specifically increase crystallinity and also affects the ionic conductivity of the films. This study reports on the morphological and chemical properties of the bio-composite films fabricated with 50% keratin and 50% cellulose, coagulated in hydrogen peroxide by analyzing Fourier Transform Infrared Spectroscopy (FTIR), Thermogravimetric Analysis (TGA), Differential Scanning Calorimetric (DCS), and Scanning Electron Microscopy (SEM).

NIDHI SHETH, GRGICAK LAB

TITLE: Developing a forensically relevant single-cell pipeline for human identification

ABSTRACT: Interpreting mixture samples from a bulk pipeline is arduous since the signal is a cacophony of low-fidelity fluorescence from noise, artifact and incomplete allele signal from anunknown number of contributors. The resultant electropherograms (EPGs) are sometimes so complex as to require significant computational power to complete the interpretation. An alternate to the bulk-processing pipeline is a single-cell one, where the sample is collected and each cell is sequestered. The DNA from each cell is then extracted, amplified and electrophoresed. The result is n single-source single-cell EPGs from n cells. In the single-cell pipeline, an efficient direct-to-PCR treatment that is compatible with all forensically relevant downstream processes is necessary.

In the first part of the study four direct-to-PCR extraction treatments (forensicGEM®Saliva (ZyGEM Corp Ltd); DEPArray™ LysePrep Kit (Menarini Silicon Biosystems); Direct PCR Lysis Reagent (Viagen Biotech, Inc.); and Arcturus® PicoPure™ DNA Extraction Kit (Arcturus Bioscience, Inc.)) were tested on 408 buccal cells procured with a low throughput single-cell sequestration method — i.e., Pico pipetting. We explored if allele dropout rates are cell-dependentand examined whether the extraction treatment had downstream ramifications on signal quality. Signal quality was examined by interrogating four EPG metrics: allele detection rates; peak heights; peak height ratios; and peak height sloping across low to high molecular weight STR markers. The results show that, overall, 77% of the cells (313 out of 408 single–cells) rendered EPGs with at least a 50% allele detection rate, and that allele drop-out rates were cell dependent. Permutation tests demonstrated that extraction treatments significantly impacted all metrics of EPG quality. Notably, the Arcturus® PicoPureTM extraction treatment resulted in the lowest median allele drop-out rate, the highest median average peak height, the highest median average peak height ratio, and lowest median EPG sloping values.

In the second part of the study we replace the manual pico-pipetting method with a semi-automated one by integrating micro-fluidics and DEPArrayTM technology. Phosphate buffer saline (PBS) is an important reagent in this process, as it mitigates cell rupture. If the cells rupture prior to DNA extraction it could result in poor signal quality, negatively affectingforensic outcomes. Despite PBS’s value in mitigating cell-rupture, it is a known PCR inhibitor. It, therefore, became necessary to optimize PBS concentrations to ensure pre-mature cell lysis was minimized while maximizing PCR efficiency. Thus, we use the same metrics of EPG quality to test the effects of four different PBS+Pro K extraction treatments on 241 single leukocytes: 1. High Pro K/ 1X PBS; 2. Low Pro K/1X PBS 3. Low Pro K/0.5X PBS; and Low Pro K/0.25X PBS. By decreasing the PBS concentration to 0.5X the highest overall signal quality was obtained. The results also indicated that lowering the Pro K concentration did not significantly impact signal quality. We therefore conclude that a lower than recommended Pro K concentration during PicoPureTM direct-to-PCR extraction coupled with a lower than recommended PBS concentration is a viable treatment for semi-automated single-cell pipelines.

In addition to examining effects of candidate laboratory treatments, we shall confirm whether the signal quality is unchanged between manual and semi-automated processes while exploring if signal quality is independent of cell-type. Acquiring a systems understanding of the single-cell treatment will, therefore, inform the development of a viable single-cell systemhaving forensic relevance.

CRESTON SINGER, SALAS LAB

TITLE: Fabrication of Advanced Self-folding Agarose Materials

ABSTRACT: The alteration of materials to improve their properties can elicit unique behaviors applicable in technologies such as foldable touch screens, therapeutic delivery systems, and even novel architectural material. Of particular interest are materials obtained via renewable natural sources that have minimal environmental impact. Natural sources typically include plant products like cellulose, agar, carrageenan, and gum. These compounds are crucial for structural support and mobility in plants and can thus be modified for similar external uses. Alterations of such materials can result in even more intriguing characteristics like electrical and thermal conductivity as well as increased elasticity, plasticity, and durability. This work aims to explore different ways to improve agarose hydrogel films. Agarose is the component of agar that is responsible for the high strength gelling properties of sea plants. Through the addition of a plasticizing agent and stabilizing agent, we were able to produce an agarose film that was self-folding and could withstand repeated bending, folding, being soaked in water, and dried without retaining any permanent damage. By varying the plasticizer and stabilizing agents, in addition to creating a bilayer with the addition of silk or cellulose films, we believe we can control and tune our agarose film’s durability and folding nature. To understand the structural morphology of our agarose film, various analytical techniques were implemented including Fourier transform infrared spectroscopy (FTIR), thermal gravimetric analysis (TGA), and X-Ray scattering.

CHRISTOPHER SOTTOLANO, YAKOBY LAB

TITLE: Nebulous without white: annotated long-read genome assembly and CRISPR/Cas9 genome engineering in D. nebulosa

ABSTRACT: The plethora of morphological diversity among Drosophila species presents an opportunity to study the mechanisms underlying the molecular evolution of tissue patterning and morphologies. One of the challenges in investigating these species is that, unlike the tremendous number of molecular and genetics tools available for D. melanogaster research, many other species do not have sequenced genomes to take advantage of recent advances in said tools. Here, we focus on D. nebulosa due to the ease of rearing, its differences in tissue patterning and eggshell morphologies, and unique mating behavior when compared to D. melanogaster. As a first step, we generated a high-quality annotated genome assembly of D. nebulosa, using PacBio long-read sequencing. We used our new assembly to design an experiment that successfully disrupted the white gene via CRISPR/Cas9, but we were unable to integrate the Cas9 gene using homology-directed repair. Interestingly, unlike many other Drosophila species, D. nebulosa males null for white did not appear to court females. We conclude that gene disruption via CRISPR/Cas9 genome engineering is a useful tool in D. nebulosa, however, a selectable marker which does not impair vision is needed to replace the commonly used white-eye phenotype. The new genome assembly and annotation for this species we report here broadly enables this and other future gene editing work in D. nebulosa.

CODY STEVENS, YAKOBY LAB

TITLE: Out of the shadows: Co-acting cis¬-regulatory elements control T-box transcription factors midline and H15 during development.

ABSTRACT: Tissue development involves highly coordinated expression of transcription factors controlled by complex cis-regulatory elements (CREs). The posterior end of the follicular epithelium is patterned by Midline (MID), the Drosophila homolog of the mammalian Tbx-20 transcription factor. Here, we identify two CREs that recapitulate the endogenous pattern of mid. It has been genetically shown that mid is negatively regulated by the Iroquois transcription factor mirror and the bone morphogenetic protein signaling pathway, while positively activated by the ETS transcription factor pointed. In this study, we corroborate these trans regulators through cis regulation of mid, highlighting regulatory regions within the CREs. Furthermore, through CRISPR/Cas9 genome editing, we illustrate a redundancy of mid CREs. However, deletion of both CREs resulted in a significant reduction, but not loss of mid, indicating additional CREs are possible. Interestingly, unlike the published transregulatory connection between MID and a second Tbx-20 homolog H15, we found co-regulation between one mid CRE and H15, whereas deletion of this CRE eliminated H15 transcription. These results suggest that both T-box transcription factors mid and H15 share regulatory capacity that may act as safeguards for determination of posterior fate during egg development.