<p>Building a robust deep neural network (DNN) framework turns out to be a very difficult task as adaptive attacks are developed that break a robust DNN strategy. In this work we first study the bootstrap distribution of DNN weights and biases. We bootstrap three DNN models: a simple three layer convolutional neural network (CNN), VGG16 with 13 convolutional layers and 3 fully connected layers, and Inception v3 with 42 layers. Both VGG16 and Inception v3 are trained on CIFAR10 in order for bootstrapping networks to converge. We then compare the bootstrap NN parameter distributions with those from training DNN with different random initial seeds. We discover that the bootstrap DNN parameter distributions change as the DNN model size increases. And the bootstrap DNN parameter distributions are very close to those obtained from training with different random initial seeds. The bootstrap DNN parameter distributions are used to create a graybox defense strategy. We randomize a certain percentage of the weights of the first convolutional layers of a DNN model, and create a random ensemble of DNNs. Based on one trained DNN, we have infinitely many random DNN ensembles. The adaptive attacks lose the target. A random DNN ensemble is resilient to the adversarial attacks and maintains performance on clean data.</p>

Synopsis My research work is focused on: (i) development of novel Fast NMR methods in solution state and their application to metabolomics and small molecules. (ii) NMR based metabolic study of human IVF to assess embryo viability for implantation. The major components of the embryo growth media were identified for evaluating the embryo quality. Described below are the projects carried out towards the dissertation of my PhD. Chapter 1 describes NMR methods which are the foundation stones for new Fast NMR methods developed. Typical 1D and 2D NMR experiments used in metabolomics and statistical methods for analysis are described. A few applications of metabolomics are also covered in the chapter. Chapter 2 describes a new Fast NMR method based on polarization sharing and parallel acquisition using the dual receiver system. The method developed helps in acquiring simultaneously three 2D NMR spectra: 2D [13C-1H] HETCOR, 2D [1H-1H] TOCSY and 2D [13C-1H] HSQC-TOCSY in a single data set. This method achieves a time saving of about two fold. All the experiments are acquired on molecules with natural abundance of 13C. The method was used to assign the side chain atoms (1H and 13C) of two important peptides. i) 12 amino acid residue peptide, which is a part of central linker domain of Human Insulin like Growth Factor Binding Protein-2 known to play a vital role in the IGF system and ii) a 18 amino acid residue peptide which acts as an antimicrobial agent. Chapter 3 describes extension of the Fast NMR method described in chapter 2. The method is combined with G-matrix Fourier Transform NMR spectroscopy. In this method we have acquire simultaneously two 2D NMR experiments and one reduced dimensional 3D experiment. The three experiments are 2D [13C-1H] HETCOR, 2D [1H-1H] TOCSY and GFT (3,2)D [13C-1H] HSQC-TOCSY, which provide complementary information for rapid assignments. GFT (3,2)D [13C-1H] HSQC-TOCSY gives 3D correlations in a 2D manner facilitating high resolution and unambiguous assignments. The experiments were applied for complete assignment of 21 unlabeled metabolite mixtures corresponding to the Innovative Sequential medium (ISM1) used for culturing human embryos for IVF. Further, a 13C multiplicity edition block is added to the method to simplify the resonances assignment in GFT (3,2)D [13C-1H] HSQC-TOCSY. Taken together, experiments provide time gain of order of magnitudes compared to conventional data acquisition. Chapter 4 of the thesis describes a metabolomics study of Human in-vitro fertilization to assess viable embryos of implantation potential using NMR as non-invasive tool. NMR study included the analysis of 127 embryo culture media (Innovative Sequential Media-1) and 29 controls (culture media without embryo) of both day-2 and day-3 transferred. The embryos were divided into 3 categories 1) implanted (successful) 2) transferred not-implanted (unsuccessful) 3) not transferred based on morphological studies. All NMR experiments were acquired with CPMG (T2 filter) incorporated in 1D 1H presaturation pulse scheme. The study was based on estimation of lactate, pyruvate and alanine levels in the embryo culture media (ISM1). The study reveals higher uptake of pyruvate and high pyruvate/alanine ratios in case of implanted embryos compared to one which failed to implant. Present study provides pyruvate/alanine ratio as a biomarker to select the embryos with high implantation potential. The method combined with morphology based assessment or with other biomarkers can be serve as a powerful tool to assess the embryo quality. Chapter 5 describes a novel NMR method for rapid characterization of translation diffusion of molecules in solution either in mixture or pure form. Unlike acquisition of several 2D [13C-1H] HSQC experiments with varying gradients to get diffusion measurement, a single 2D [13C-1H] HSQC is sufficient to measure the diffusion coefficients which is in the linewidths of peaks. The method uses the idea of accordion NMR spectroscopy, wherein gradients are linearly co-incremented with 13C chemical shift evolution period during t1. The methodology speeds up the acquisition by replacing series of 2D [13C-1H] HSQC with single 2D constant time [13C-1H] HSQC. The method was used to monitor the diffusion of metabolites in a time-resolved manner during polymerization of SDS-PAGE gel. Using this method, it was possible to detect the presence of oligomers of diphenylalanine (FF) during its self assembly to form nanotubular structures.

Metabolomics is the study of metabolites and metabolic processes. Due to the diversity of structures and polarities of metabolites, no single analytical technique is able to measure the entire metabolome - instead a varied set of experimental designs and instrumental technologies are used to measure specific portions. This has led to the development of many distinct data analysis and processing methods and software. There is hope that metabolomics can be utilized for clinical applications, in toxicology and to measure the exposome. However, for these applications to be realised data must be high quality, sufficiently standardised and annotated, and FAIR (Findable, Accessible, Interoperable and Reproducible). For this purpose, it is also important that standardised, FAIR software workflows are available. There has also recently been much concern over the reproducibility of scientific research, which FAIR and open data, and workflows can help to address. To this end, this thesis aims to assess current practices and standards of sharing data within the field of metabolomics, using metascientific approaches. The types of functions of software for processing and analysing metabolomics data is also assessed. Reporting standards are designed to ensure that the minimum information required to un- derstand and interpret the results of analysis are reported. However, poor reporting standards are ignored and not complied with. Compliance to the biological context Metabolomics Standards Initiative (MSI) guidelines was examined, in order to investigate their timeliness. The state of open data within the metabolomics community was examined by investigating how much publicly available metabolomics data there is and where has it been deposited. To explore whether journal data sharing policies are driving open metabolomics data, which journals publish articles that have their underlying data made open was also examined. However, open data alone is not inherently useful: if data is incomplete, lacking in quality or missing crucial metadata, it is not valuable. Conversely, if data are reused, this can demonstrate the worth of public data archiving. Levels of reuse of public metabolomics data were therefore examined. With greater than 250 software tools specific for metabolomics, practitioners are faced with a daunting task to select the best tools for data collection and analysis. To help educate researchers about what software is available, a taxonomy of metabolomics software tools and a GitHub pages wiki, which provides extensive details about all included software, have been developed.

Ein genereller Ansatz zur Charakterisierung von biologischen Systemen bietet die Untersuchung des Metaboloms, dessen Analyse als „Metabolomics“ bezeichnet wird. “Omics”- Technologien haben das Ziel, ohne Selektionskriterien möglichst alle Bestandteile einer biologischen Probe zu detektieren (identifizieren und quantifizieren), um daraus Rückschlüsse auf nicht vorhersehbare und somit neuartige Korrelationen in biologischen Systemen zu ziehen. Ein zentrales Dogma in der Biologie besteht in der Kausalität zwischen Gen – Enzym – Metabolite. Perturbationen auf einer Ebene rufen systemische Antworten hervor, die in einem veränderten Phänotyp münden können. Metabolite sind die Endprodukte von zellulären regulatorischen Prozessen, deren Abundanz durch die Resonanz auf genetische Modifikationen oder Umwelteinflüsse zurückzuführen ist. Zudem repräsentieren Metabolite ultimativ den Phänotyp eines Organismus und haben die Fähigkeit als Biomarker zu fungieren. Die integrale Analyse verschiedenster Stoffwechselwegen wie Krebszyklus, Pentosephosphatzyklus oder Calvinzyklus offeriert die Identifikation von metabolischen Mustern. In dieser Arbeit wurden sowohl das targeted Profiling via GC-TOF-MS als auch das untargeted Profiling via GC-TOF-MS und LC-FT-MS als analytische Strategien genutzt, um biologische Systeme anhand ihrer Metabolite zu charakterisieren und um physiologische Muster als Resonanz auf endogene oder exogene Stimuli zu erkennen. Dabei standen die metabolische, phänotypische und genotypische Plastizität von Pflanzen im Fokus der Untersuchungen. Metabolische Varianzen eines Phänotyps reflektieren die genotyp-abhängige Resonanz des Organismus auf umweltbedingte Parameter (abiotischer und biotischer Stress, Entwicklung) und können mit sensitiven Metabolite Profiling Methoden determiniert werden. Diese Anwendungen haben unter anderem auch zum Begriff des „Environmental Metabolomics“ geführt. In Kapitel 2 wurde der Einfluss biotischer Interaktionen von endophytischen Bakterien auf den Metabolismus von Pappelklonen untersucht; Kapitel 3 betrachtet die metabolische Plastizität von Pflanzen im Freiland auf veränderte biotische Interaktionsmuster (Konkurrenz/Diversität/Artenzusammensetzung); Abschließend wurde in Kapitel 4 der Einfluss von spezifischen genetischen Modifikationen an Peroxisomen und den daraus resultierenden veränderten metabolischen Fluss der Photorespiration dargestellt. Aufgrund der sensitiven Analyse- Technik konnten metabolische Phänotypen, die nicht zwingend in einen morphologischen Phänotyp mündeten, in drei biologischen Systemen identifiziert und in einen stoffwechselphysiologischen Kontext gestellt werden. Die drei untersuchten biologischen Systeme – in vitro- Pappeln, Grünland- Arten (Arrhenatherion-Gesellschaft) und der Modellorganismus (Arabidopsis) – belegten anschaulich die Plastizität des Metabolismus der Arten, welche durch endogene oder exogene Faktoren erzeugt wurden. / A general approach to characterise biological systems offers the analysis of the metabolome, named “metabolomics”. “Omics”- technologies are untargeted approaches without any selection criteria which aim to detect every potential analyte in a sample in order to draw conclusions about new correlations in biological systems. A central dogma in biology is the causality between gene – enzyme – metabolite. Perturbations on one level are reflected in systemic response, which possibly result in a changed phenotype. Metabolites are end products of its gene expression and metabolism, whose abundance is determined as a resonance of genetic modifications or environmental disturbance. Furthermore metabolites represent the ultimate phenotype of an organism and are able to act as a biomarker. The integral analysis of distinct metabolic pathways like TCA, Pentose phosphate and Calvin cycle consequently leads to the identification of metabolic patterns. In this work targeted profiling via GC-TOF-MS as well as untargeted profiling via GC-TOF-MS and LC-FT-MS were used as analytical strategies to characterise biological systems on the basis of their metabolites and to identify physiological patterns as resonance of endogenic or exogenic stimuli. The focus of the investigations concentrates on the metabolic, phenotypic and genotypic plasticity of plants. Metabolic variance of a phenotype is reflected in the genotypic dependence response of an organism on environmental parameters which may be detected via sensitive metabolic profiling methods. In chapter 2 the influence of biotic interaction of endophytic bacteria on the metabolism of their poplar host was analyzed; chapter 3 explores the metabolic plasticity of field-grown grassland species as a consequence of biotic interaction pattern (competition / diversity / species composition); In conclusion, chapter 4 illustrates the influence of specific genetic modifications on peroxisomes and the consequent changed metabolic flux in the photorespiration pathway. Due to the sensitive analytic methods, metabolic phenotypes in all three biological systems could be identified and classified in a physiological context. The three biological systems – in vitro poplar plants, field-grown grassland species and the model organism Arabidopsis – demonstrate the plasticity of the metabolism of species in response to stimuli.

There is an ever increasing need of new technologies for identification of molecular markers for early diagnosis of fatal diseases to allow efficient treatment. In addition, there is great value in finding patterns of metabolites, proteins or genes altered in relation to specific disease conditions to gain a deeper understanding of the underlying mechanisms of disease development. If successful, scientific achievements in this field could apart from early diagnosis lead to development of new drugs, treatments or preventions for many serious diseases.  Metabolites are low molecular weight compounds involved in the chemical reactions taking place in the cells of living organisms to uphold life, i.e. metabolism. The research field of metabolomics investigates the relationship between metabolite alterations and biochemical mechanisms, e.g. disease processes. To understand these associations hundreds of metabolites present in a sample are quantified using sensitive bioanalytical techniques. In this way a unique chemical fingerprint is obtained for each sample, providing an instant picture of the current state of the studied system. This fingerprint or picture can then be utilized for the discovery of biomarkers or biomarker patterns of biological and clinical relevance. In this thesis the focus is set on evaluation and application of strategies for studying metabolic alterations in human tissues associated with disease. A chemometric methodology for processing and modeling of gas chromatography-mass spectrometry (GC-MS) based metabolomics data, is designed for developing predictive systems for generation of representative data, validation and result verification, diagnosis and screening of large sample sets. The developed strategies were specifically applied for identification of metabolite markers and metabolic pathways associated with prostate cancer disease progression. The long-term goal was to detect new sensitive diagnostic/prognostic markers, which ultimately could be used to differentiate between indolent and aggressive tumors at diagnosis and thus aid in the development of personalized treatments. Our main finding so far is the detection of high levels of cholesterol in prostate cancer bone metastases. This in combination with previously presented results suggests cholesterol as a potentially interesting therapeutic target for advanced prostate cancer. Furthermore we detected metabolic alterations in plasma associated with metastasis development. These results were further explored in prospective samples attempting to verify some of the identified metabolites as potential prognostic markers.

Quantitative Untersuchungen krankheitsspezifischer oder krankheitsassoziierter metabolischer Signaturen in humanen Körperflüssigkeiten („Clinical Metabolomics“) haben zum Ziel neue Ansätze für diagnostische oder therapeutische Konzepte zu entwickeln. Die simultane quantitative Analytik von Aminosäuren (AS) und Acylcarnitinen (AC) mittels Tandem-Massenspektrometrie (MS/MS) ermöglicht die Erfassung wichtiger Stoffwechselwege des humanen Metabolismus. Hierzu zählen der Stoffwechsel der ketogenen AS, des Harnstoffzyklus oder der β-Oxidation langkettiger Fettsäuren. Allerdings wird die Konzentration der verschiedenen metabolischen Parameter in humanen Körperflüssigkeiten durch eine Vielzahl präanalytischer in vitro Störfaktoren und in vivo Einflussgrößen beeinflusst. Diese können zu signifikanten Veränderungen der Laborergebnisse führen. Im Rahmen meiner Promotionsarbeit wurden in vitro Störfaktoren (Probenmaterial, Lagerung u. a.) und in vivo Einflussgrößen (Ernährung, physische Aktivität) untersucht und ein standardisiertes Präanalytik-Protokoll entwickelt. Dazu wurden pro Probe 3 µL Trockenblut (TB), 10 µL Serum oder Plasma nach Butylierung mittels Elektrospray-Ionisations-MS/MS analysiert und jeweils 26 AS und 35 AC in 1,5 Minuten simultan bestimmt. Als Ergebnis der zahlreichen systematischen Präanalytik-Untersuchungen konnten signifikante Konzentrationsunterschiede der Metabolite zwischen kapillärer und venöser Blutentnahme sowie in Abhängigkeit des Hämatokrits gefunden werden. Im Vergleich zu Serum und antikoaguliertem Plasma (EDTA, Citrat, Heparin) waren die Konzentrationen der langkettigen AC im TB 5-fach höher. Nahrungsaufnahme und körperliche Aktivität führten ebenfalls zu signifikanten Veränderungen der AS- und AC-Konzentrationen. Durch Optimierung des Probenaufarbeitungsprotokolls konnte die Variabilität zwischen den Messtagen für 17 AS und 6 AC auf < 20 % gesenkt werden. Die Ergebnisse meiner Promotionsarbeit unterstreichen den Einfluss präanalytischer Faktoren auf die Metabolomanalytik. Durch Etablierung und Einhaltung standardisierter präanalytischer Protokolle kann die präanalytische Varianz der Ergebnisse deutlich verringert werden. Sie stellen somit eine wichtige Voraussetzung für eine qualitativ hochwertige Metabolomanalytik im Rahmen klinischer Studien zur Identifizierung neuer Biomarker dar.

Quantitative Untersuchungen krankheitsspezifischer oder krankheitsassoziierter metabolischer Signaturen in humanen Körperflüssigkeiten („Clinical Metabolomics“) haben zum Ziel neue Ansätze für diagnostische oder therapeutische Konzepte zu entwickeln. Die simultane quantitative Analytik von Aminosäuren (AS) und Acylcarnitinen (AC) mittels Tandem-Massenspektrometrie (MS/MS) ermöglicht die Erfassung wichtiger Stoffwechselwege des humanen Metabolismus. Hierzu zählen der Stoffwechsel der ketogenen AS, des Harnstoffzyklus oder der β-Oxidation langkettiger Fettsäuren. Allerdings wird die Konzentration der verschiedenen metabolischen Parameter in humanen Körperflüssigkeiten durch eine Vielzahl präanalytischer in vitro Störfaktoren und in vivo Einflussgrößen beeinflusst. Diese können zu signifikanten Veränderungen der Laborergebnisse führen. Im Rahmen meiner Promotionsarbeit wurden in vitro Störfaktoren (Probenmaterial, Lagerung u. a.) und in vivo Einflussgrößen (Ernährung, physische Aktivität) untersucht und ein standardisiertes Präanalytik-Protokoll entwickelt. Dazu wurden pro Probe 3 µL Trockenblut (TB), 10 µL Serum oder Plasma nach Butylierung mittels Elektrospray-Ionisations-MS/MS analysiert und jeweils 26 AS und 35 AC in 1,5 Minuten simultan bestimmt. Als Ergebnis der zahlreichen systematischen Präanalytik-Untersuchungen konnten signifikante Konzentrationsunterschiede der Metabolite zwischen kapillärer und venöser Blutentnahme sowie in Abhängigkeit des Hämatokrits gefunden werden. Im Vergleich zu Serum und antikoaguliertem Plasma (EDTA, Citrat, Heparin) waren die Konzentrationen der langkettigen AC im TB 5-fach höher. Nahrungsaufnahme und körperliche Aktivität führten ebenfalls zu signifikanten Veränderungen der AS- und AC-Konzentrationen. Durch Optimierung des Probenaufarbeitungsprotokolls konnte die Variabilität zwischen den Messtagen für 17 AS und 6 AC auf < 20 % gesenkt werden. Die Ergebnisse meiner Promotionsarbeit unterstreichen den Einfluss präanalytischer Faktoren auf die Metabolomanalytik. Durch Etablierung und Einhaltung standardisierter präanalytischer Protokolle kann die präanalytische Varianz der Ergebnisse deutlich verringert werden. Sie stellen somit eine wichtige Voraussetzung für eine qualitativ hochwertige Metabolomanalytik im Rahmen klinischer Studien zur Identifizierung neuer Biomarker dar.

Inositolpyrophosphate (PP-InsP) sind eine Gruppe sekundärer Signalmoleküle, die in einer Vielzahl zellulärer Prozesse, von Phosphathomeostase über Insulinsignalisierung bis Apoptose eine Rolle spielen. Die Art und Weise, wie PP-InsPs ihre Funktion ausführen, noch weitgehend unbekannt. Deshalb wurden zwei neue analytische Methoden basierend auf Kernspinresonanzspektroskopie und Flüssigchromatographie mit Massenspektrometrie-Kopplung (LCMS) entwickelt. Um die limitierende Sensitivität der Kernresonanzspektroskopie zu umgehen, wurde die Synthese von kernspinresonanzaktivem, 13C-markiertem Inositol optimiert. Des Weiteren wurde eine chemoenzymatische Synthese für alle Säugetier-PP-InsP-Isomere entwickelt, die auf der skalierbaren Ausfällung mittels Mg2+ Ionen basiert. Menschliche Zellen wurden mit 13C-Inositol isotopenmarkiert und in den Spektren der Zellextrakte wurde, basierend auf den PP-InsP-Standards, Fingerabdrucksignale identifiziert mit denen die Konzentrationen der dazugehörigen Moleküle bestimmt werden konnte. Die LCMS basierte Methode wurde auf dem Prinzip der Umsetzung von hochgeladenen Inositolpyrophosphaten zu ihren korrespondieren Methylestern mittels Trimethylsilyldiazomethan geplant. Die ungeladenen, permethylierten PP-InsPs wären geeignet für LC-Auftrennungen und MS-Messungen und sollten eine von Kernspinresonanzspektroskopie nicht erreichbare Sensitivität ermöglichen. Die Methode wurde mittels Inositolhexakisphosphat (InsP6), einem einfacheren PP-InsP-Analog, etabliert und methyliertes InsP6 konnte in Mengen von 10 femtomol detektiert werden. Die Adaption der Methode für die PP-InsPs gestaltete sich jedoch herausfordernd, da der Analyt während der Reaktion zersetzt wurde. Ein Wechsel zu Diazomethan als Methylierungsagens zeigte vielversprechende Resultate. / Inositol pyrophosphates (PP-InsPs) are a well conserved group of second messengers that are involved in a plethora of cellular processes including phosphate homeostasis, insulin signaling, and apoptosis. Despite much effort, it is still mostly unknown how PP-InsPs exert their diverse functions. In order to decipher the mechanisms, researchers have relied either on metabolic labeling with radioactive inositol or on electrophoretic separation on polyacrylamide gels but these methods either lack ease of use or sensitivity. Therefore, two new analytical tools, based on nuclear magnetic resonance (NMR) spectroscopy, and liquid chromatography coupled mass spectrometry (LCMS), were developed. To overcome the limited sensitivity provided by NMR spectroscopy, a high yielding synthesis of NMR-active 13C-labeled inositol was designed and optimized. Furthermore, a chemoenzymatic synthesis of all mammalian PP-InsPs isomers was developed that relied on a scalable purification strategy utilizing precipitation with Mg2+ ions. Human cells were metabolically labeled with 13C-inositol and the prepared PP-InsPs were used as standards to identify peaks in the NMRspectra. These fingerprint signals enabled the quantification of the corresponding molecules. The LCMS-based method was based on the derivatization of the highly charged inositol pyrophosphates to their corresponding methyl esters by trimethylsilyldiazomethane. The permethylated InsPs and PP-InsPs were suitable for LC separation and MS measurement, and provide a sensitivity unmatched by NMR spectroscopy. The method was established using inositol hexakisphosphate, a simpler analog of PP-InsPs, and methylated InsP6 could be detected at quantities as low as 10 femtomole. However, the adaptation of the derivatization for PP-InsPs proved challenging as the reaction caused degradation of the analyte but strategies to circumvent the decay by changing the derivatization agent to diazomethane were promising.

BACKGROUND:The rhizome, the original stem of land plants, enables species to invade new territory and is a critical component of perenniality, especially in grasses. Red rice (Oryza longistaminata) is a perennial wild rice species with many valuable traits that could be used to improve cultivated rice cultivars, including rhizomatousness, disease resistance and drought tolerance. Despite these features, little is known about the molecular mechanisms that contribute to rhizome growth, development and function in this plant.RESULTS:We used an integrated approach to compare the transcriptome, proteome and metabolome of the rhizome to other tissues of red rice. 116 Gb of transcriptome sequence was obtained from various tissues and used to identify rhizome-specific and preferentially expressed genes, including transcription factors and hormone metabolism and stress response-related genes. Proteomics and metabolomics approaches identified 41 proteins and more than 100 primary metabolites and plant hormones with rhizome preferential accumulation. Of particular interest was the identification of a large number of gene transcripts from Magnaportha oryzae, the fungus that causes rice blast disease in cultivated rice, even though the red rice plants showed no sign of disease.CONCLUSIONS:A significant set of genes, proteins and metabolites appear to be specifically or preferentially expressed in the rhizome of O. longistaminata. The presence of M. oryzae gene transcripts at a high level in apparently healthy plants suggests that red rice is resistant to this pathogen, and may be able to provide genes to cultivated rice that will enable resistance to rice blast disease.

<p>Gliomas are tumors of the central nervous system for which improvements in treatment are critically needed. Mutations in IDH1 and IDH2, which encode the cytosolic and mitochondrial NADP+-dependent isocitrate dehydrogenases, respectively, are frequent in gliomas. Here, we summarize recent literature concerning gliomas, the normal cellular functions of IDH1/2, the epidemiology of IDH1/2 mutations, and the understanding of the function of IDH1/2 mutations in cancer. We then show in vitro using liquid chromatography-mass spectrometry that a function of many IDH1/2 mutations is to produce 2-hydroxyglutarate. Next, we use a mass spectrometry based platform to characterize metabolic changes in a glioma cell line expressing IDH1/2 mutants and show that the IDH mutants are associated with lowered N-acetylated amino acids both in this cell line model and in primary tumor tissue. Finally, we develop and characterize a Drosophila melanogaster (fruit fly) model of IDH1/2-mutated cancer by expressing the mutated Drosophila homolog of IDH1 in fly tissues using the UAS-Gal4 binary expression system. These results delineate downstream molecular players that likely play a role in IDH1/2-mutated cancer and provide a model organism for interrogation of genetic networks that interact with IDH1/2 mutation. These findings refine our understanding of glioma pathogenesis and may inform the design of new glioma therapies.</p> / Dissertation