Organizational Unit:

School of Biological Sciences

Permanent Link

https://hdl.handle.net/1853/70750

Parent Organization

Organizational Unit

College of Sciences

Includes Organization(s)

Organizational Unit

Center for the Study of Systems Biology

ArchiveSpace Name Record

https://finding-aids.library.gatech.edu/agents/corporate_entities/1131

Full item page

Publication Search Results

Now showing 1 - 10 of 130

Porewater Constituents Serve As A Control And Sensitive Indicator Of The Belowground Carbon Cycle Response To Climate Change Drivers

(Georgia Institute of Technology, 2022-12-14) Song, Tianze

Although comprising only ~3% of the terrestrial surface area, peatland ecosystems store an estimated one-third or more of global soil carbon. In addition to this role as a net carbon sink, freshwater wetlands, such as peatlands, account for approximately one-third of global methane (CH4) emissions to the atmosphere. While CH4 emissions from peatlands are expected to disproportionately increase due to warming, the environmental controls remain poorly constrained. Any increase in CH4 emission is of great concern due to the fact that the sustained-flux global warming potential of CH4 is estimated to be 34-times greater than that of carbon dioxide (CO2) on a 100 y timescale. The majority of heterotrophic organic matter decomposition and the production of greenhouse gases in peatlands occurs in the soils, which are composed largely of solid degraded peat and interstitial waters or porewaters. Microorganisms in peatlands soils reside attached to the solid peat particles or are free-living in the planktonic porewater environment. Dissolved organic matter (DOM) in porewaters represents a primary driver of heterotrophic respiration and methanogenesis in surficial peat soils across many peatland types. Evidence also indicates that DOM can inhibit anaerobic decomposition and CH4 production by serving as a source of alternative electron acceptors, acidifying the environment, or releasing anti-microbial polyphenolic compounds, which calls for further mechanistic understanding of porewater-peat interactions. To this end, this dissertation employed laboratory and field studies to investigate the response of peat and porewater microbial communities to climate change factors (warming, elevated atmospheric CO2). In Chapter 2, the response of soil organic matter decomposition and greenhouse gas (GHG) production to climate drivers was investigated in a laboratory study of peat soils collected from a temperate bog dominated by peat mosses (S1 bog) at the Marcell Experiment Forest in northern Minnesota. Amendment of peat soil microcosms with porewater inhibited GHG production and contributed to acetate pooling at close to in situ temperature (4°C) but stimulated GHG production and acetate consumption rates by up to a factor of 2 at warmer temperatures (14 and 25°C). Elevated temperature (25°C) led to a slight decrease in microbial diversity and stimulated the growth of methanogens and specific syntrophic taxa. These results confirm that DOM is a primary driver of decomposition in peatland soils, contains compounds that inhibit microbial metabolism, and show that inhibition is alleviated by warming. In Chapter 3, porewater microbial communities were characterized in peatlands for the first time and compared to their counterparts attached to solid phase peat soil. In addition, given that soil core sampling can be tedious and destructive, especially in the context of whole ecosystem climate manipulation experiments, the feasibility of using porewater microbial communities as an indicator of ecosystem response to climate drivers (warming and elevated atmospheric CO2) was explored. This portion of my research leveraged the whole ecosystem warming (WEW) experiment entitled, Spruce and Peatland Responses Under Changing Environments (SPRUCE) located in a northern peatland. Results revealed that planktonic microbial communities in porewaters are distinct from attached communities associated with the solid phase of peat soil. Microbial abundance was two to three orders of magnitude lower in porewater compared to peat, whereas porewater communities contained a substantially higher alpha diversity compared to peat. Pronounced shifts in the abundance, diversity, and composition of porewater microbial communities were observed with warming. In contrast, attached peat microbial communities showed a relatively muted response to increasing temperature, although microbial abundance did increase with warming at the surface. Specific functional guilds of microorganisms that mediate the methane cycle (methanotrophs) were shown to increase in relative abundance with increasing temperature. While porewater microbial communities show promise as a sensitive indicator of climate change, further investigation is warranted to elucidate the metabolic potential of predominant taxa in porewaters and whether the response of porewater community dynamics to climate drivers is representative of overall peat soil community dynamics.
Computational Analysis Methodologies for Evaluating Metabolism Changes In iPSCs Undergoing Differentiation

(Georgia Institute of Technology, 2022-12-14) Nikitina, Arina Antonovna

Induced pluripotent stem cells (iPSCs) hold great promise as a regenerative medicine tool, allowing the creation of any type of tissue from a patient’s own cells. Clinical applications of iPSCs are still hampered by the great costs of cell manufacturing, as well as the poor understanding of the processes governing the differentiation process. Additionally, few non-invasive algorithms have been established to evaluate quality control at the earliest stage of differentiation before lengthy protocols expend resources. To identify quality control attributes and enhance knowledge of iPSC differentiation, I developed a multi-modal imaging analytical pipeline that resolves spatial metabolomics to single-cell resolution. In this work, I investigated changes in iPSC lipid profiles during the initial loss of pluripotency over the course of spontaneous differentiation using co-registration of confocal microscopy and matrix-assisted laser desorption/ionization (MALDI) mass spectrometry imaging. Despite lipids having known functions in cell signaling, their role in pluripotency maintenance and lineage specification is underexplored. Lipids that are highly informative of the temporal stage of the differentiation were identified through a variety of multivariate modeling methods and shown to reveal lineage bifurcation occurring metabolically. Among these lipids, several phosphatidylinositol (PI) species emerged as early metabolic markers of pluripotency loss, preceding changes in Oct4 - a transcription factor well known for its connection to pluripotency. In addition, continuous inhibition of phosphatidylethanolamine N-methyltransferase during differentiation enhanced pluripotency maintenance and increased levels of the same lipid pluripotency markers before transcriptional changes. This highlights a novel mechanism of pluripotency maintenance by upregulating PI production via an unknown pathway. The small subset of informative PI species can be used as novel targets for early quality control in a variety of protocols involving pluripotency loss and/or maintenance. In addition, I investigated the predictive power of real-time oxygen consumption measurements in iPSCs undergoing cardiomyocyte differentiation through machine learning algorithms. Oxygen consumption rate values in the first 4 days of the differentiation protocol were highly predictive of the cardiomyocyte yield by day 16, with several time-series features capable of predicting the differentiation failure as early as in the first 48 hours. This approach can be easily scaled and translated to the regenerative medicine application, greatly reducing cell manufacturing costs by saving time and resources from being wasted on a failed batch. In summary, this work presents new analytical tools for cellular manufacturing that also yields novel knowledge about iPSCs metabolism during differentiation.
Revealing patterns of ribonucleotide incorporation in genomic DNA of eukaryotic cells

(Georgia Institute of Technology, 2022-12-13) Xu, Penghao

Ribonucleoside triphosphates (rNTPs) are incorporated into DNA by DNA polymerases in the form of ribonucleoside monophosphates (rNMPs), which are the most abundant modified nucleotides in genomic DNA in many species. The presence of rNMPs in DNA leads to DNA structural change, and genome instability, affects protein-DNA interactions, and results in diseases including Aicardi-Goutières Syndrome. The incorporated rNTPs are mainly removed by the ribonucleotide excision repair (RER) pathway initiated by Ribonuclease (RNase) H2. Previous researchers already developed several rNMP-mapping techniques to capture the rNMPs in DNA at single nucleotide level. However, the analyses of rNTP incorporation are still limited to the basic study of rNMP counts and compositions in small genomes of RNase H2-deactivated cells. In this study, we first developed a series of key bioinformatics tools to maximize the efficiency of the rNMP-mapping procedure, and to uncover unique features of rNMP presence in the genomic DNA of cells. We optimized the restriction enzyme usage of rNMP-mapping techniques to significantly increase the rNMP capture efficiency, which allowed us to perform detailed rNMP-pattern analysis and reveal correlations between genomic rNMP sites and specific genetic elements. The development of heatmaps showing the composition and sequence context of the rNMPs in genomic DNA allowed us to discover preferred rNTP-incorporation patterns and hotspot motifs in the full genome of many eukaryotic cells including budding and fission yeast, green algae, and human cells. By studying incorporated rNTP patterns around the origins of replication in S. cerevisiae, we revealed rNMP-pattern changes associated with the DNA replication process, which validated the contribution of DNA Pol δ in the early stage of nascent leading strand synthesis. Moreover, we discovered a strong preference for rNMPs after dAMPs on the leading strands and after dCMPs on the lagging strands, which are the specific rNTP incorporation patterns of eukaryotic replicative DNA Pol ε and Pol δ, respectively. These signatures of rNMPs in DNA can be used to track the DNA polymerase usage in DNA replication across the yeast genome and possibly in other eukaryotic cells. Finally, by studying rNTP-incorporation patterns in human mitochondrial DNA (mtDNA), we found a strong light strand preference for rNTP-incorporation and rNMP-enriched zones in human mtDNA. Three consistent rNMP-enriched zones, found within the replication control region of human mtDNA, may affect the activity of DNA Pol γ and that of TWINKLE, the main helicase of human mtDNA. We also showed that longer coding sequences in human mtDNA have higher rNTP-incorporation frequency on their non-template strands at each base, which revealed the direct association between rNMPs and genes in human mtDNA.
Differential Gene Co-Expression Network Characteristics Of Cancer

(Georgia Institute of Technology, 2022-12-13) Arshad, Zainab

The transformation from a healthy state to a disease state in cancer is dictated in large part by structural and regulatory abnormalities in genes. While the molecular features underlying this transition have been investigated for some time, allowing groundbreaking advancements in cancer research, a majority of these efforts are focused on mutational and expression changes of individual genes. The recent advancement of network-based analytic methods affords an additional route through which disease pathophysiology and biologic regulation can be investigated. Furthermore, with the development of high-throughput technologies and the availability of large biobanks, gene interaction changes, and their functional consequences can be reliably interpreted from a systemic perspective, in a context specific manner. Towards this end, my research investigates gene co-expression changes, derived from transcriptomic case-control data, that underlie cancer onset and progression relative to healthy tissue. For the first study, global network changes associated with cancers of nine different tissues of origin were investigated. Network complexity generally dropped in the transition from normal precursor tissues to corresponding primary tumors, whereas cross-tissue cancer network similarity overall increased in early-stage cancers followed by a subsequent loss in similarity as tumors reacquire cancer-specific network complexity in late-stage cancers. In addition, gene-gene connections remaining stable through cancer development were found enriched for ‘‘housekeeping’’ gene functions, whereas newly acquired interactions were associated with established cancer-promoting functions. For the second study, gene-network characteristics of the molecular subtypes (Luminal A, Luminal B and Basal) of Breast Cancer (BC) were outlined based on a comparative analysis relative to precursor normal breast tissue. Basal was identified as the most highly connected yet dissimilar subtype to normal control. We discovered eight extensively connected network modules acquired in Basal BCs that harbored 19 genes found significantly associated with survival and encoding cancer hallmark functions including regulation of cell proliferation and motility, as well as neural pathways that have not been previously associated with basal BCs. Finally, the consensus approach of network construction for an unbiased differential analysis of gene co-expression networks used in these studies was published as a step-by-step protocol. Altogether, this thesis highlights gene-network changes characteristic of individual cancer types, molecular subtypes and disease stages that informs their diverse progression patterns and clinical outcomes. Furthermore, it underscores the importance and demonstrates the utility of gene co-expression networks in identifying key genes, gene interactions and functional characteristics of cancers that maybe undiscovered by standard molecular analysis approaches.
R-Pyocin Regulation, Release, and Susceptibility in Pseudomonas aeruginosa

(Georgia Institute of Technology, 2022-12-09) Mei, Madeline

Pseudomonas aeruginosa is a Gram-negative opportunistic pathogen and a major determinant of declining lung function in individuals with cystic fibrosis (CF). P. aeruginosa possesses many intrinsic antibiotic resistance mechanisms and isolates from chronic CF lung infections develop increasing resistance to multiple antibiotics over time. Chronic infection with P. aeruginosa remains one of the main causes of mortality and morbidity in CF patients, thus new therapeutic interventions are necessary. R-type pyocins are narrow spectrum, phage tail-like bacteriocins, specifically produced by P. aeruginosa to kill other strains of P. aeruginosa. Due to their specific anti-pseudomonal activity and similarity to bacteriophage, R-pyocins have potential as additional therapeutics for P. aeruginosa, either in isolation, in combination with antibiotics, or as an alternative to phage therapy. There are five subtypes of R-pyocin (types R1-R5), and it is thought that each P. aeruginosa strain uniquely produces only one of these, suggesting a degree of strain-specificity. P. aeruginosa from CF lung infections develop increasing resistance to antibiotics, making new treatment approaches essential. It is known P. aeruginosa populations in CF chronic lung infection become phenotypically and genotypically diverse over time, however, little is known of the efficacy of R-pyocins against heterogeneous populations. Even less is known regarding the timing and regulation of R-pyocins in CF lung infections, or if P. aeruginosa utilizes R-pyocin production during infection for competition or otherwise – which may influence pressure towards R-pyocin resistance. In this work, I evaluated R-pyocin type and susceptibility among P. aeruginosa isolates sourced from CF infections and found that (i) R1-pyocins are the most prevalent R-type among respiratory infection and CF strains; (ii) a large proportion of P. aeruginosa strains lack R-pyocin genes entirely; (iii) isolates from P. aeruginosa populations collected from the same patient at a single time point have the same R-pyocin type; (iv) there is heterogeneity in susceptibility to R-pyocins within P. aeruginosa populations and (v) susceptibility is likely driven by diversity of LPS phenotypes within clinical populations. These findings suggest that there is likely heterogeneity in response to other types of LPS-binding antimicrobials, including phage, which is important for consideration of antimicrobials as therapeutics. To investigate the prevalence of R2-pyocin susceptible strains in CF, I then utilized 110 isolates of P. aeruginosa collected from five individuals with CF to test for R2-pyocin susceptibility and identify LPS phenotypes. From our collection we i) estimated that approximately 83% of sputum samples contain heterogenous P. aeruginosa populations without R2-pyocin resistant isolates and all sputum samples contained susceptible isolates; ii) we found that there is no correlation between R2-pyocin susceptibility and LPS phenotypes, and iii) we estimate that approximately 76% of isolates sampled from sputum lack O-specific antigen, 42% lack common antigen, and 27% exhibit altered LPS cores. This finding highlights that perhaps LPS packing density may play a more influential role in mediating R-pyocin susceptibility in infection. Finding the majority of our sampled P. aeruginosa populations to be R2-pyocin susceptible further supports the potential of these narrow-spectrum antimicrobials despite facing heterogenous susceptibility among diverse populations. In order to evaluate how R-pyocins may influence strain competition and growth in CF lung infection, I assessed R-pyocin activity in an infection-relevant environment (Synthetic Cystic Fibrosis Sputum Medium; SCFM2) and found that (i) R-pyocins genes are transcribed more in the CF nutrient environment than in rich laboratory medium and (ii) in a structured, CF-like environment, R-pyocin induction is costly to producing strains in competition rather than beneficial. Our work suggests that R-pyocins may not be essential in CF lung infection and can be costly to producing cells in the presence of stress response-inducing stimuli, such as those commonly found in infection. In this thesis I have studied R-pyocin susceptibility, regulation and release utilizing a biobank of whole populations of P. aeruginosa collected from 11 individuals with CF, as well as the CF infection model (SCFM) to understand the mechanisms of R-pyocin activity in an infection-relevant context and the role R-pyocins play in shaping P. aeruginosa populations during infection. The findings of this work have illuminated the impact of P. aeruginosa heterogeneity on R-pyocin susceptibility, furthered our understanding of R-pyocins as potential therapeutics, and built upon our knowledge of bacteriocin-mediated interactions.
Ecological determinants of ecosystem stability

(Georgia Institute of Technology, 2022-12-08) Xu, Qianna

Understanding the mechanisms underpinning the stability of ecological systems has been a central goal of community ecology. This dissertation presents studies examining the roles of several important ecological factors, including species diversity, nutrient enrichment, and interspecific competition, in regulating ecological stability. First, I conducted comprehensive quantitative meta-analyses that synthesized the findings of existing empirical studies of the relationships between species diversity and temporal stability of ecosystem properties. The meta-analysis revealed a general, consistently positive relationship between species diversity and ecosystem temporal stability, a robust finding that holds even after accounting for the confounding effects of environmental covariates. This study also identified species compensatory dynamics as an important mechanism underlying the observed positive diversity-ecosystem stability relationships. Second, I investigated the effects of nutrient enrichment on multiple dimensions of compositional and functional stability, using semi-arid grasslands as the study system. I found that nitrogen enrichment reduced grassland functional and compositional temporal stability, resistance, and recovery, but increased grassland functional and compositional resilience. Importantly, nitrogen enrichment influenced most functional stability dimensions by altering their corresponding compositional stability dimensions, whereas reduced species diversity under nitrogen enrichment contributed little to observed changes in grassland stability. Lastly, using laboratory microcosms containing assemblages of freshwater bacterivorous protozoans as model systems, I examined the effects of interspecific competition and determinants of competitive outcomes (species niche and fitness differences) on ecosystem temporal stability. I found that the presence of competition increased species asynchrony but not population stability, resulting in more stable ecosystem dynamics. Species niche differences increased species asynchrony, whereas species fitness differences decreased population stability, leading to changes in ecosystem temporal stability. These results demonstrate the linkage between species coexistence mechanisms and ecological stability, providing a mechanistic understanding of competition effects on population and ecosystem stability. Altogether, this dissertation contributes novel knowledge to the field of ecological stability, with important implications for the stable provisioning of ecosystem products and services under ongoing global environmental change.
Anthropogenic-Mediated Simplification of Marine Food Webs

(Georgia Institute of Technology, 2022-12-08) Willert, Madison Shari

Anthropogenic-mediated stressors such as overexploitation, habitat destruction, climate change, and species introductions are changing food webs in marine ecosystems. In this dissertation, I first evaluate how these stressors are shifting trophic interactions via increased dietary overlap and interspecific competition within trophic levels, truncation of nutrient flow between trophic levels and ecosystems, and simplification and compression of entire food webs. Stable isotope analysis is a powerful tool to measure species’ trophic positions and thus, food web shifts over time and space. I show that δ15N values and δ13C values from formalin-preserved seaweeds are generally reliable, validating stable isotope analysis of herbarium specimens. Seaweeds are useful as nutrient baselines for trophic ecology studies, as well as for assessing nutrient runoff and pollution; this finding shows that preserved herbarium specimens can be used in these types of studies to reconstruct food webs of the past. I then use nitrogen stable isotope analysis of both herbarium specimens and museum fish specimens from New England, USA to show that the common piscivore Centropristis striata (black sea bass) and the common benthivore Stenotomus chrysops (scup) have experienced significant declines in trophic position in this area since pre-1950. Centropristis striata declined almost a full trophic level and Stenotomus chrysops declined half a trophic level, and these species are now converging on similar trophic positions coincident with the increase in destructive bottom fishing in New England. Next, I used nitrogen stable isotope analysis of >1000 museum fish specimens from coral reefs worldwide to assess dietary changes of common coral reef mesopredators since 1850 in regions of both the tropical Atlantic and the Indo-Pacific. I found that trophic instability has been common in the tropical Atlantic during the 20th century, with the trophic position of most Atlantic species decreasing further going into the 21st century. Unlike in the Atlantic, historically unstable species in the Indo-Pacific are now increasing in their trophic positions; this suggests that relatively higher levels of overfishing and coral loss in the tropical Atlantic are reflected in greater mesopredator trophic instability. Finally, I used nitrogen and carbon stable isotope analysis of vertebrae from Sphyrna mokarran (great hammerhead) and Sphyrna lewini (scalloped hammerhead) sharks to evaluate ontogenetic shifts in these two species in the U.S. South Atlantic and the eastern Gulf of Mexico. Sphyrna lewini occupies a high trophic position throughout its life, reaching peak predator status as a subadult and occupying more offshore pelagic habitats. Despite its larger body size, Sphyrna mokarran occupies a lower trophic position and relies more on benthic and inshore habitats, especially in the juvenile stage. I elucidated the nuances of these predators’ trophic ecology and found no evidence of within-species differences in sex or location with regards to dietary habits. A better understanding of individual species’ trophic ecology, as well as historic human impacts on marine food webs, is crucial to maintaining and promoting healthy ecosystems into the future.
Evolution of Cell Differentiation and Whole Genome Duplication in a Multicellularity Long-Term Evolution Experiment

(Georgia Institute of Technology, 2022-12-01) Tong, Kai

The evolution of multicellular organisms from unicellular ancestors is considered a major transition in the history of life, but significant gaps remain in our understanding of how multicellularity originates and evolves. Over the last decade, experimental evolution has emerged as a powerful approach to study multicellular evolution in real time. Our lab has been conducting a Multicellularity Long-Term Evolution Experiment (MuLTEE), evolving the simple cluster-forming yeast Saccharomyces cerevisiae (snowflake yeast) with daily selection for larger size for 1000 days (~5000 generations). In my thesis, I used the MuLTEE to explore two widespread and impactful phenomena in the evolution of multicellularity: the evolution of cell differentiation, and whole genome duplication. First, using single-cell RNA sequencing, I identified the evolution of three cell states within individual clusters, which differentially upregulate ribosomal processes, mitochondrial gene expression and cell wall biogenesis, and stress response and apoptosis, respectively. These cell states display conserved gene expression signatures over evolution as well as isolate-specific features, and present interesting evolutionary dynamics characterized by the broad emergence and expansion of cellular heterogeneity with isolate-specific losses. Second, using quantitative microscopy and whole-genome sequencing, I revealed the convergent early emergence and long-term maintenance of tetraploidy from diploid ancestors, which is in striking contrast with previous studies in yeast where diploidy is stable while nascent tetraploidy rapidly reverts to diploidy. A synthetic construction and reversion experiment showed that under selection for larger size, tetraploidy provides an immediate fitness advantage, by producing larger, more elongated cells that yield larger clusters, while evolved tetraploids rapidly underwent ploidy reduction in the absence of size selection. Taken together, these results shed new light on how cell differentiation can evolve from simple groups of cells, and how the emergence and maintenance of polyploidy can be driven by continuous selection on its immediate phenotypic effects, further demonstrating the power of long-term experimental evolution in understanding the processes driving multicellular evolution.
Nitrate and Phosphate Loads, but not Light Availability, Impact Freshwater Phytoplankton Diversity via Tradeoffs Between Dominant Species

(Georgia Institute of Technology, 2022-12-01) Southard, Michael Kelton

Eutrophication of freshwater ecosystems, mainly caused by nitrogen (N) and phosphorus (P) pollution, causes significant economic damages every year in the U.S. Excess N and P deposition in lakes can result in harmful algal blooms, reduced biodiversity, and increased greenhouse gas production, but we still do not fully understand how and why phytoplankton communities react to nutrient enrichment under varying conditions. Several theories – including the niche dimension hypothesis, biomass-driven competition hypothesis, nitrogen detriment hypothesis, and benthic model – are currently being explored in both terrestrial and aquatic producer communities in attempt to better understand the biological mechanisms effecting these systems, and the goal of this study was to determine which models are most applicable to freshwater phytoplankton. Using five-species microcosms of green algae, we found that N and P enrichment significantly reduced diversity (independent of light availability), which was likely mediated through nutrient tradeoffs between the two dominant species, Ankistrodesmus falcatus and Selenastrum capricornutum. Additionally, we observed a significant decrease in monoculture carrying capacity across all species with high N addition in low P concentrations, indicating that high N:P ratios may be physiologically harmful to green algae. These findings suggest that the niche dimension and nitrogen detriment hypotheses may be the most applicable to freshwater phytoplankton communities and could be useful for protecting and mitigating economic losses from these systems.
Genetic, Molecular, and Environmental Effects on Survival Outcomes and Disparities

(Georgia Institute of Technology, 2022-11-21) Lee, Kara Keun

While disease mortality rates have steadily decreased over the last century, survival disparities among racial and ethnic population groups persist. Research on population survival disparities tends to be focused on specific diseases, with self-identified race and ethnicity (SIRE) and genetic ancestry (GA) often used interchangeably for studying genetic and environmental effects on disparities. Moreover, the underlying multidimensional factors and mechanisms that drive survival disparities are still largely unknown. This thesis explores the genetic, molecular, and environmental effects on overall and cause-specific survival disparities across population groups by leveraging genetic ancestry inference and population biobank data. For the first study, cancers with significant survival disparities between GA and SIRE groups were identified using The Cancer Genome Atlas (TCGA), followed by the characterization of differential molecular signatures that interact with GA and exacerbate cancer survival disparities (CSD). For the second study, two of the four cancers showing significant CSD between African and European GA groups were further characterized by constructing GA-specific gene co-expression networks, revealing targeted sets of genes and associated biological processes underlying CSD in African ancestry patients. Finally, the third study broadens the research scope beyond cancer to investigate the overall and leading cause-specific mortality disparities in the UK Biobank. It shifts focus to environmental risk factors contributing to disparities in mortality across different ethnic groups. Numerous blood biomarkers and modifiable environmental and behavioral mortality risk factors were identified, several of which showed differential effects on mortality across ethnic groups. Altogether, this thesis highlights (1) the group-specific genetic and environmental risk factors that contribute to disparities in survival outcomes, and (2) the utility and importance of population-specific study designs that leverage ancestry information and integrative analysis frameworks, combining clinical outcome data with genomic and socioenvironmental data, for greater clinical relevance and translatability of research findings that can help to improve health equity.