Rapidly accumulating genomic data as a result of high-throughput sequencing has necessitated development of efficient computational methods to decode the biological information underlying these data. DNA composition varies across structurally or functionally different regions of a genome as well as those of distinct evolutionary origins. We adapted an integrative framework that combines a top-down, recursive segmentation algorithm with a bottom-up, agglomerative clustering algorithm to decipher compositionally distinct regions in genomes. The recursive segmentation procedure entails fragmenting a genome into compositionally distinct segments within a statistical hypothesis testing framework. This is followed by an agglomerative clustering procedure to group compositionally similar segments within the same framework. One of our main objectives was to decipher distinctive evolutionary patterns in sex chromosomes via unraveling the underlying compositional heterogeneity. Application of this approach to the human X-chromosome provided novel insights into the stratification of the X chromosome as a consequence of punctuated recombination suppressions between the X and Y from the distal long arm to the distal short arm. Novel "evolutionary strata" were identified particularly in the X conserved region (XCR) that is not amenable to the X-Y comparative analysis due to massive loss of the Y gametologs following recombination cessation. Our compositional based approach could circumvent the limitations of the current methods that depend on X-Y (or Z-W for ZW sex determination system) comparisons by deciphering the stratification even if only the sequence of sex chromosome in the homogametic sex (i.e. X or Z chromosome) is available. These studies were extended to the plant sex chromosomes which are known to have a number of evolutionary strata that formed at the initial stage of their evolution, presenting an opportunity to examine the onset of stratum formation on the sex chromosomes. Further applications included detection of horizontally acquired DNAs in extremophilic eukaryote, Galdieria sulphuraria, which encode variety of potentially adaptive functions, and in the taxonomic profiling of metagenomic sequences. Finally, we discussed how the Markovian segmentation and clustering method can be made more sensitive and robust for further applications in biological and biomedical sciences in future.

The arrival of genome sequencing technologies has triggered a transformative surge in genomic data, revolutionizing the ability to study the genome's role in biological processes. Despite this progress, understanding how genomic elements work together to sustain life remains a critical challenge. In this work, we develop computational frameworks that bridge gaps in current genome modeling approaches by integrating biologically meaningful inductive biases. First, we reformulate genome-wide association studies to account for nonlinear modes of variant interaction. We supplement this analysis by building polygenic risk scores to explore the interplay between genetics and environmental factors contributing to phenotype variation. The second part of this thesis shifts focus to deep learning approaches for improved sequence modeling. We present a novel application of hyperbolic convolutional neural networks to exploit the evolutionarily-informed structure of sequence data, enabling more expressive DNA sequence representations. We demonstrate how our class of models discern phylogenetic structure amidst noisy signals, and further motivate our work by constructing an empirical method for interpreting the hyperbolicity of dataset embeddings. Next, we leverage state space models to study an instance of long range genome interaction in the form of topologically associating domains. Our framework accurately recapitulates known patterns of chromatin organization, providing insights into epigenetic regulation. Altogether, these methods contribute to the development of advanced computational methods that align with the challenges of genomic sequence modeling and pave the way for a more comprehensive understanding of genome organization and function.

Functional genomic screens using CRISPR technology have revolutionized the discovery of genetic vulnerabilities and synthetic lethal interactions in cancer, including hematological malignancies. Our work aims to contribute to existing knowledge on the molecular underpinnings governing response to targeted therapies in acute leukemia by leveraging the power of genome-wide CRISPR screens. Specifically, we studied T cell acute lymphoblastic leukemia (T-ALL) and post-myeloproliferative neoplasm acute myeloid leukemia (post-MPN AML) which are aggressive hematological conditions with dismal prognosis in relapsed and advanced-age patients, respectively. Therapeutic targeting of NOTCH1 signaling in T-ALL using gamma secretase inhibitors (GSIs) and JAK/STAT signaling in post-MPN AML using Type I JAK2 inhibitors have shown promising results but were limited due to modest antileukemic activity and adverse side effects. Thus, finding synthetic lethal targets that could enhance the therapeutic activities of GSIs and JAK2 inhibitors are warranted. Here, we performed genome-wide CRISPR screens to discover genes that show synergistic interactions with GSI and with JAK2 inhibitors. We present potential therapeutic targets to improve the antileukemic effects of GSI in T-ALL. In addition, we show that inhibition of the neddylation pathway circumvents GSI-induced gut toxicity by stabilizing HES1 protein. Lastly, we demonstrate that CREBBP/EP300 bromodomain inhibition potentiates JAK2 inhibition in post-MPN AML by downregulating oncogenic MYC and phosphorylated STATs levels and activity. Collectively, our findings corroborate the power of genome-wide CRISPR screens integrated with downstream functional and multi-omics analysis for the identification and characterization of potential therapeutic targets in hematological malignancies.

The recovery of metagenome-assembled genomes is biased towards the most abundant species in a given community. To improve the identification of species, even if only dominant species are recovered, we investigated the integration of flow cytometry cell sorting with bioinformatics tools to recover metagenome-assembled genomes. We used a cell culture of a wastewater microbial community as our model system. Cells were separated based on fluorescence signals via flow cytometry cell sorting into sub-communities: dominant gates, low abundant gates, and outer gates into subsets of the original community. Metagenome sequencing was performed for all groups. The unsorted community was used as control. We recovered a total of 24 metagenome-assembled genomes (MAGs) representing 11 species-level genome operational taxonomic units (gOTUs). In addition, 57 ribosomal operational taxonomic units (rOTUs) affiliated with 29 taxa at species level were reconstructed from metagenomic libraries. Our approach suggests a two-fold increase in the resolution when comparing sorted and unsorted communities. Our results also indicate that species abundance is one determinant of genome recovery from metagenomes as we can recover taxa in the sorted libraries that are not present in the unsorted community. In conclusion, a combination of cell sorting and metagenomics allows the recovery of MAGs undetected without cell sorting.

No description available.

Indiana University-Purdue University Indianapolis (IUPUI) / My thesis comprises of two individual projects: 1) we have developed a database for operon prediction using high-throughput sequencing datasets for bacterial genomes. 2) Genomics and mechanistic insights of convergent transcription in bacterial genomes. In the first project we developed a database for the prediction of operons for bacterial genomes using RNA-seq datasets, we predicted operons for bacterial genomes. RNA-seq datasets with different condition for each bacterial genome were taken into account and predicted operons using Rockhopper. We took RNA-seq datasets from NCBI with distinct experimental conditions for each bacterial genome into account and analyzed using tool for operon prediction. Currently our database contains 9 bacterial organisms for which we predicted operons. User interface is simple and easy to use, in terms of visualization, downloading and querying of data. In our database user can browse through reference genome, genes present in that genome and operons predicted from different RNA-seq datasets. Further in the second project, we studied the genomic and mechanistic insights of convergent transcription in bacterial genomes. We know that convergent gene pairs with overlapping head-to-head configuration are widely spread across both eukaryotic and prokaryotic genomes. They are believed to contribute to the regulation of genes at both transcriptional and post-transcriptional levels, although factors contributing to their abundance across genomes and mechanistic basis for their prevalence are poorly understood. In this study, we explore the role of various factors contributing to convergent overlapping transcription in bacterial genomes. Our analysis shows that the proportion of convergent overlapping gene pairs (COGPs) in a genome is affected due to endospore formation, bacterial habitat, oxygen requirement, GC content and the temperature range. In particular, we show that bacterial genomes thriving in specialized habitats, such as thermophiles, exhibit a high proportion of COGPs. Our results also conclude that the density distribution of COGPs across the genomes is high for shorter overlaps with increased conservation of distances for decreasing overlaps. Our study further reveals that COGPs frequently contain stop codon overlaps with the middle base position exhibiting mismatches between complementary strands. Further, for the functional analysis using cluster of orthologous groups (COGs) annotations suggested that cell motility, cell metabolism, storage and cell signaling are enriched among COGPs, suggesting their role in processes beyond regulation. Our analysis provides genomic insights into this unappreciated regulatory phenomenon, allowing a refined understanding of their contribution to bacterial phenotypes.

Die Endprodukte der Tetrapyrrolbiosynthese sind essentiell für die Schwefel- und Stickstoffassimilation (Sirohäm), der von Photorezeptoren abhängigen Genexpression (Phytochromobilin), Elektronenübertragungsreaktionen (Häm) und der Photosynthese (Chlorophyll). Die Synthese von Chlorophyllen wird durch eine Mg-Chelatase (MgCh) eingeleitet, die durch das GENOMES UNCOUPLED 4 (GUN4) Protein stimuliert wird. GUN4 ist essentiell für die Aktivierung der MgCh und die Synthese von Chlorophyllen. Das GUN4 aus Arabidopsis thaliana wird ausschließlich an der vorletzten Aminosäure (S264) des C-Terminus phosphoryliert. Die in vitro und in vivo MgCh-Aktivität wird hingegen durch phosphoryliertes GUN4 nicht mehr stimuliert. De-phosphoryliertes GUN4 bewirkt die lichtabhängige Aktivierung der MgCh im Übergang von der Nacht zum Tag in Angiospermen. Im Laufe der Evolution photosynthetisch aktiver Organismen hat sich die in den Angiospermen hochkonservierte Phosphorylierungsstelle entwickelt. GUN4-Homologe aus Synechocystis oder Chlamydomonas werden nicht phosphoryliert. Im Rahmen der Suche nach der GUN4-spezifischen Proteinkinase wurden vier in den Plastiden lokalisierte PLASTID PROTEIN KINASE WITH UNKNOWN FUNCTION identifiziert. In dieser Arbeit wurden zusätzlich Experimente zum durch die GUN-Proteine vermittelten retrograden Signalweg durchgeführt. gun Mutanten sind durch eine defizitäre cytosolische Anthocyan-/Flavonoidbiosynthese charakterisiert. Auf der Suche nach Hinweisen für einen Zusammenhang zwischen Anthocyanen und der De-repression von PHOTOSYNTHESIS-ASSOCIATED NUCLEAR GENES wurde eine neue gun Mutante identifiziert. Der knockout der durch TRANSPARENT TESTA 4 (TT4) kodierten CHALCON SYNTHASE führte zu einer mit den gun Mutanten vergleichbaren De-repression der PHANGs nach Norflurazon-Behandlung. Pharmakologische Experimente belegen eine mögliche Funktion der Phenylpropanoidbiosynthese in der durch die GUN-Proteine vermittelten retrograden Kommunikation. / Endproducts of the tetrapyrrole biosynthesis pathway are essential for the assimilation of sulfur and nitrogen (siroheme), photoreceptor mediated control of nuclear gene expression (phytochromobilin), electron transfer reactions (heme) and photosynthesis (chlorophyll). The synthesis of chlorophyll is initiated by a Mg-chelatase (MgCh) which is stimulated by the GUN4 protein. GUN4 is essential for the activation of MgCh and synthesis of chlorophyll. GUN4 from Arabidopsis thaliana is exclusively phosphorylated at the next-to-last amino acid of the C-terminus (S264). The stimulatory impact towards MgCh is reduced upon GUN4 phosphorylation. De-phosphorylated GUN4 stimulates MgCh activity during the transition from night to daytime. The phosphorylation site of GUN4 has evolved in the clade of angiosperms. GUN4 homologs of Synechocystis or Chlamydomonas are not phosphorylated. In an attempt to isolate the GUN4-kinase four formerly unknown PLASTID PROTEIN KINASE WITH UNKNOWN FUNCTION were identified. In addition to the elucidation of the post-translational GUN4 modifications, experiments concerning the GUN-dependent retrograde signaling pathway were performed. Under conditions which lead to a block of chloroplast development the de-repression of PHOTOSYNTHESIS-ASSOCIATED NUCLEAR GENES is paralleled by a reduced accumulation of anthocyanins in the gun mutants. When searching for a correlation between anthocyanin biosynthesis and expression of PHANGs a new gun mutant was identified. The knockout of CHALCONE SYNTHASE encoded by TRANSPARENT TESTA 4 (TT4) leads to a comparable de-repression of PHANGs after norflurazon treatment as it was observed for the gun mutants. Pharmacological modification of phenylpropanoid biosynthesis revealed that an intermediate of the pathway is a component of chloroplast-to-nucleus communication. Hence, first evidences for a function of the phenylpropanoid biosynthesis pathway in mediating the GUN-dependent retrograde signal were obtained.

Oenococcus oeni joue un rôle essentiel dans l’élaboration du vin. Adaptée aux environnements acides et riches en alcool, elle est la bactérie lactique naturellement sélectionnée pour mener la fermentation malolactique (FML). Elle est ainsi la principale espèce recherchée et utilisée industriellement comme levain malolactique. Toutefois, il existe une grande diversité phénotypique au sein des souches d’O. oeni et notamment une variabilité des propriétés technologiques que sont la résistance à la lyophilisation, la résistance à l’inoculation dans le vin et la capacité à réaliser rapidement la FML. De nombreux gènes impliqués dans l’adaptation au vin ont déjà été identifiés mais, ne se sont pas toujours révélés efficaces pour la sélection de souches œnologiques. Dans ce contexte, cette étude a consisté à identifier des gènes spécifiques des souches d’intérêt technologique à travers l’analyse des plasmides et génomes. Face aux difficultés rencontrées pour purifier les grands plasmides, seul le plasmide pOENI-1 a été étudié. Ce travail a révélé différentes formes plasmidiques regroupées en une famille nommée « pOENI-1 ». Plusieurs gènes accessoires ont été identifiés et deux d’entre eux ont été détectés chez les souches associées aux fermentations malolactiques spontanées. La comparaison des génomes de souches aux propriétés technologiques diverses a également révélé des séquences génétiques qui leur sont spécifiques. L’ensemble de ces travaux a permis d’identifier plusieurs gènes dont la distribution statistique parmi les souches d’O. oeni a été analysée par la construction de courbes ROC. Ces courbes permettent d’évaluer la qualité des gènes en tant que marqueurs génétiques des souches d’intérêt technologique. Il est donc maintenant possible d’orienter la sélection des nouveaux levains malolactiques par l’utilisation des données génétiques et des outils statistiques décrits dans cette étude. / Oenococcus oeni plays an essential role in the production of wine. Adapted to acidic and alcohol rich environments, it is the lactic acid bacterium species that is naturally selected to conduct malolactic fermentation (MLF). It is also the main species that is selected and used industrially as malolactic starter. However, there is a huge phenotypic diversity among strains of O. oeni, which includes a variability of technological properties such as resistance to freeze-drying, resistance to inoculation into the wine and the ability to quickly achieve the MLF. Many genes involved in adaptation to wine have been identified but have not always proven effective in selecting wine strains. In this context, this study aimed to identify genes that are specific strains of technological interest through the analysis of genomes and plasmids. Due to difficulties encountered to purify large plasmids, only the plasmid pOENI-1 was studied. This work has revealed several different but related plasmids that were grouped into a family named "pOENI-1". Several accessory genes have been identified and two of them were detected in O. oeni strains associated with spontaneous MLF. Comparing the genomes of strains showing various technological properties also revealed genetic sequences that are specific of those strains. Altogether, these works have revealed several genes whose statistical distribution among O. oeni strains was analyzed by constructing ROC curves. These curves are used to assess the quality of genes as genetic markers of strains of technological interest. It is now possible to guide the selection of new malolactic starters by the use of genetic data and statistical tools described in this study.

A recombinação durante a transcrição reversa é um fator importante no aumento da diversidade genética e adaptação do HIV-1, permitindo que mutações vantajosas presentes em diferentes linhagens sejam combinadas em um mesmo genoma. No Brasil, vários recombinantes foram descritos e seis formas recombinantes circulantes (CRFs) já foram identificados, demonstrando a relevância destes recombinantes na epidemia brasileira. Portanto, um dos objetivos desta tese foi analisar os dados gerados pela Rede de Diversidade Genética Viral (VGDN) (sequências parciais de gag, pol e env), a fim de identificar recombinantes inter-subtipos de HIV-1 e avaliar a frequência e distribuição geográfica destes vírus. Utilizando diferentes técnicas foram identificados 152/1083 pacientes portadores de recombinantes BF. A frequência destes recombinantes foi maior em cidades como São Vicente (30%) e Sorocaba (22,6%), sendo que os recombinantes circulantes em São Vicente foram geralmente relacionados às CRF28 e CRF29, enquanto que os vírus presentes na região de Sorocaba comumente apresentam um envelope subtipo F1, independente do subtipo nos demais genes. Além disso, o gene da integrase de 159 pacientes foi amplificado e sequenciado. A análise deste gene revelou mais 10 pacientes infectados com recombinantes BF e nenhuma mutação de resistência primária aos inibidores da integrase foi encontrada. O segundo objetivo foi isolar e caracterizar recombinantes BF in vitro. O isolamento viral foi realizado por co-cultivo e ao final foram obtidos 10 isolados primários. O sequenciamento do genoma quase completo desses dez isolados primários revelou que três isolados primários pertencem ao grupo da CRF28_BF, três ao grupo da CRF29_BF e quatro foram classificados como formas recombinantes únicas (URFs). Ainda, o uso de correceptores desses isolados foi avaliado in vitro em ensaios com as células GHOST(3), e revelou três duplo-trópicos (X4/R5) vírus, quatro CXCR4 (X4) e três isolados utilizaram apenas CCR5 (R5). Em suma, uma alta frequência de URFs foi encontrada em algumas cidades do Estado de São Paulo, e também foi desenvolvido e caracterizado um painel de isolados primários representando as CRF28_BF, CRF29_BF e algumas URFs. / Recombination during reverse transcription is an important factor promoting HIV-1 diversity and adaptive change, allowing advantageous mutations arising on different genomes to undergo linkage in the same progeny recombinant genome more frequently than what would be expected under random mutation alone. In Brazil, several recombinant viruses were reported, and six circulating recombinant forms (CRFs) have already been identified. Therefore, the first objective of this Thesis was to analyze the data generated by the Viral Genetic Diversity Network (VGDN) (gag, pol and env partial sequences), in order to identify HIV-1 intersubtype recombinants and evaluate the frequency and geographical distribution of these viruses. Using different techniques we identified 152/1083 patients harboring BF recombinants. The frequency of these recombinants was higher in cities like São Vicente (30%) and Sorocaba (22.6 %). The recombinant viruses circulating in São Vicente were generally related to CRF28 and CRF29, while those viruses circulating in Sorocaba commonly presented an envelope region of subtype F1, irrespective the subtype composition on the remaining genes. Additionally, the integrase gene of HIV-1 from 159 patients was further amplified and sequenced. The analysis of this viral gene revealed ten more patients infected with BF recombinants and no primary mutations related to integrase inhibitor resistance were found. The second objective was to isolate and characterize BF recombinants in vitro, which resulted in ten primary HIV-1 isolates. The near full-length genomes of these ten primary isolates revealed that three were related to CRF28_BF, three to CRF29_BF and four were unique recombinant forms (URFs), according to their breakpoints profile determined with the jpHMM program. Additionally, the coreceptor usage of these isolate was investigated in vitro using GHOST assays, which revealed three dual-tropic (X4/R5) viruses, four CXCR4 (X4) viruses and three CCR5 (R5) viruses. In sum, we report a high frequency of URFs in some cities of São Paulo State, and also developed a well-characterized panel of viruses representing CRF28_BF, CRF29_BF and URFs.

O objetivo deste trabalho foi caracterizar uma nova família de retrotransposons Copia em cana-de-açúcar, e também compreender a diversidade de retrotransposons com LTR transcricionalmente ativos em cana, com base na análise filogenética de cDNAs provenientes do projeto SUCEST. Foi assim isolada uma nova família de retrotransposons de cana, denominada SURE (SUgarcane REtrotransposon). O genoma da cana apresentou grande colinearidade com os genomas de sorgo e arroz em segmentos contendo SURE. Foram também classificados filogeneticamente cDNAs e seqüências completas de retrotransposons com LTR de cana-de-açúcar. A maioria das linhagens evolutivas descritas apresenta ao menos um representante em cana. O grupo de retrotransposons com LTR mais representado no transcriptoma de cana foi reunido como um novo membro da superfamília Copia Garapa. Trata-se de um elemento com mais cópias e maior variabiliadade que SURE. Dessa forma, SURE e Garapa apresentam-se como duas linhagens de retrotransposons que apresentam padrões distintos de amplificação no genoma de cana. / The aim of this work was to characterize a sugarcane element belonging to a new Copia retrotransposon-family in sugarcane and also study the diversity of transcriptional active LTR-retrotransposons in sugarcane, based in a phylogenetic analysis of cDNAs from SUCEST. The new retrotransposon family identified in sugarcane was named SURE (Sugarcane REtrotransposon). BAC clones bearing one copy of SURE showed that sugarcane genome have a high sinteny with sorghum and rice genomes in coding regions. Other sugarcane sequences containing LTR-retroelements were characterized and compared to pre existent in the plant kingdom. Most of these lineages had at least one sugarcane element. The most representative group of these sequences was reunited as a new group of Copia superfamily in sugarcane and named Garapa. It comprises elements highly distributed in sugarcane chromosomes and with a higher variability when compared to SURE. Finally, SURE and Garapa represent two distinct lineages of Copia retrotransposons that had different amplification pattern in the sugarcane genome.