Non-coding RNAs (ncRNAs) are a subset of RNAs that do not code for protein. They are divided into a number of different groups based on their function and targets. Small nucleolar RNAs (snoRNAs) are ncRNAs that have long been known to function as guides for ribosomal RNA (rRNA) modifying enzymes. They are classified into two major groups: box C/D snoRNAs and box H/ACA snoRNAs. Most box C/D snoRNAs direct the 2'-O-methylation of rRNA substrates, but some lack known targets and are therefore termed 'orphan snoRNAs'. Studies have implicated orphan snoRNAs in pre-mRNA processing and stability, but the functional consequence of snoRNA binding to mRNAs has not been fully determined. Saccharomyces cerevisiae had two orphan snoRNAs, snR4 and snR45, with no known function in ribosome synthesis. This project aimed to determine the targets of these snoRNAs, and investigate the effects of snoRNA binding to non-canonical target RNAs, as well as the underlying mechanism. Synthetic gene array screens with deletions of the SNR4 and SNR45 genes identified multiple positive and negative genetic interactions. In particular, deletion of either snoRNA gene was synthetic-lethal with mutation of the snoRNA-associated methyltransferase, Nop1 (Fibrillarin in humans), demonstrating that both have important functions. CLASH analyses of RNA-RNA interactions showed that these snoRNAs bind multiple mRNAs, while RNA sequencing and RT-qPCR revealed that snoRNA deletion altered mRNA abundance. Both orphan snoRNAs were well conserved between fungi, with a region of high conservation indicating a potential binding site. Associations were identified between snR4 and snR45 and multiple sequences within rRNA, including two recently identified sites of 18S rRNA acetylation. Work elsewhere showed that snR4 and snR45 function as guides for the acetyltransferase Kre33 using the region of high conservation, removing their 'orphan' status. Orphan snoRNAs have been implicated in human diseases, such as Prader Willi Syndrome and cancers. The work discussed in this thesis helps to elucidate the RNA interactions of yeast orphan snoRNAs. It has provided a greater understanding of the mechanisms involved, and may inform future work in combatting human disease.

The last few years have seen dramatic advances in genomics, including the discovery of a large number of non-coding and antisense transcripts. This has revolutionised our understanding of multifaceted transcript structures found within gene loci and their roles in the regulation of development, neurogenesis and other complex processes. The recent and continuing surge of knowledge has prompted researchers to reassess and further dissect gene loci. The ghrelin gene (GHRL) gives rise to preproghrelin, which in turn produces ghrelin, a 28 amino acid peptide hormone that acts via the ghrelin receptor (growth hormone secretagogue receptor/GHSR 1a). Ghrelin has many important physiological and pathophysiological roles, including the stimulation of growth hormone (GH) release, appetite regulation, and cancer development. A truncated receptor splice variant, GHSR 1b, does not bind ghrelin, but dimerises with GHSR 1a, and may act as a dominant negative receptor. The gene products of ghrelin and its receptor are frequently overexpressed in human cancer While it is well known that the ghrelin axis (ghrelin and its receptor) plays a range of important functional roles, little is known about the molecular structure and regulation of the ghrelin gene (GHRL) and ghrelin receptor gene (GHSR). This thesis reports the re-annotation of the ghrelin gene, discovery of alternative 5’ exons and transcription start sites, as well as the description of a number of novel splice variants, including isoforms with a putative signal peptide. We also describe the discovery and characterisation of a ghrelin antisense gene (GHRLOS), and the discovery and expression of a ghrelin receptor (growth hormone secretagogue receptor/GHSR) antisense gene (GHSR-OS). We have identified numerous ghrelin-derived transcripts, including variants with extended 5' untranslated regions and putative secreted obestatin and C-ghrelin transcripts. These transcripts initiate from novel first exons, exon -1, exon 0 and a 5' extended 1, with multiple transcription start sites. We used comparative genomics to identify, and RT-PCR to experimentally verify, that the proximal exon 0 and 5' extended exon 1 are transcribed in the mouse ghrelin gene, which suggests the mouse and human proximal first exon architecture is conserved. We have identified numerous novel antisense transcripts in the ghrelin locus. A candidate non-coding endogenous natural antisense gene (GHRLOS) was cloned and demonstrates very low expression levels in the stomach and high levels in the thymus, testis and brain - all major tissues of non-coding RNA expression. Next, we examined if transcription occurs in the antisense orientation to the ghrelin receptor gene, GHSR. A novel gene (GHSR-OS) on the opposite strand of intron 1 of the GHSR gene was identified and characterised using strand-specific RT-PCR and rapid amplification of cDNA ends (RACE). GHSR-OS is differentially expressed and a candidate non-coding RNA gene. In summary, this study has characterised the ghrelin and ghrelin receptor loci and demonstrated natural antisense transcripts to ghrelin and its receptor. Our preliminary work shows that the ghrelin axis generates a broad and complex transcriptional repertoire. This study provides the basis for detailed functional studies of the the ghrelin and GHSR loci and future studies will be needed to further unravel the function, diagnostic and therapeutic potential of the ghrelin axis.

Aligner des macromolécules telles que des protéines, des ADN et des ARN afin de révéler ou exploiter, leur homologie fonctionnelle est un défi classique en bioinformatique, qui offre de nombreuses applications, notamment dans la modélisation de structures et l'annotation des génomes. Un certain nombre d'algorithmes et d'outils ont été proposés pour le problème d'alignement structure-séquence d'ARN. Cependant, en ce qui concerne les ARN complexes, comportant des pseudo-noeuds, des interactions multiples et des paires de bases non canoniques, de tels outils sont rarement utilisés dans la pratique, en partie à cause de leurs grandes exigences de calcul, et de leur incapacité à supporter des types généraux de structures. Récemment, Rinaudo et al. ont donné un algorithme paramétré général pour la comparaison structure-séquence d'ARN, qui est capable de prendre en entrée n'importe quel type de structures comportant des pseudo-noeuds. L'algorithme paramétré est un algorithme de programmation dynamique basée sur la décomposition arborescente. Nous avons développé plusieurs variantes et extensions de cet algorithme. Afin de l'accélérer sans perte sensible de précision, nous avons introduit une approche de programmation dynamique par bandes. De plus, trois algorithmes ont été développés pour obtenir des alignements sous-optimaux. De plus, nous introduisons dans ce contexte la notion de MEA (Maximum-expected Structure-Alignment) pour calculer un alignement avec la précision maximale attendue sur un ensemble d'alignements. Tous ces algorithmes ont été implémentés dans un logiciel nommé LiCoRNA (aLignment of Complex RNAs). Les performances de LiCoRNA ont été évaluées d'abord sur l'alignement des graines des familles de de la base de données RFAM qui comportent des pseudo-noeuds. Comparé aux autres algorithmes de l'état de l'art, LiCoRNA obtient généralement des résultats équivalents ou meilleurs que ses concurrents. Grâce à la grande précision démontrée par LiCoRNA, nous montrons que cet outil peut être utilisé pour améliorer les alignements de certaines familles de RFAM qui comportent des pseudo-noeuds. / Aligning macromolecules such as proteins, DNAs and RNAs in order to reveal, or conversely exploit, their functional homology is a classic challenge in bioinformatics, with far-reaching applications in structure modelling and genome annotation. In the specific context of complex RNAs, featuring pseudoknots, multiple interactions and non-canonical base pairs, multiple algorithmic solutions and tools have been proposed for the structure sequence alignment problem. However, such tools are seldom used in practice, due in part to their extreme computational demands, and because of their inability to support general types of structures. Recently, Rinaudo et al. gave a fully general parameterised algorithm for structure-sequence comparison, which is able to take as input any type of pseudoknotted structures. The parameterised algorithm is a tree decomposition based dynamic programming. To accelerate the dynamic programming algorithm without losing two much accuracy, we introduced a banded dynamic programming. Then three algorithms are introduced to get the suboptimal structure-sequence alignments. Furthermore, we introduce the notation Maximum Expected structure-sequence Alignment (MEA) to compute an alignment with maximum expected accuracy over a set of alignments. The Boltzmann match probability are computed based on the inside-outside algorithm. The algorithms are implemented in a software named LiCoRNA (aLignment of Complex RNAs). We first evaluate the performance of LiCoRNA on the seed alignment in the pseudoknotted RFAM families. Compared to the state-of-the-art algorithms, LiCoRNA shows generally equivalent or better results than its competitors. With the high accuracy showed by LiCoRNA, we further curate RFAM full pseudoknotted alignment. The reason why we realign full alignments is that covariance model does not support pseudoknot which may lead to misalign when building the full alignment.