With advances in high-throughput measurement techniques, large-scale biological data, such as protein-protein interaction (PPI) data, gene expression data, gene-disease association data, cellular pathway data, and so on, have been and will continue to be produced. Those data contain insightful information for understanding the mechanisms of biological systems and have been proved useful for developing new methods in disease diagnosis, disease treatment and drug design. This study focuses on two main research topics: (1) identifying protein complexes and (2) identifying disease genes from biomolecular networks. Firstly, protein complexes are groups of proteins that interact with each other at the same time and place within living cells. They are molecular entities that carry out cellular processes. The identification of protein complexes plays a primary role for understanding the organization of proteins and the mechanisms of biological systems. Many previous algorithms are designed based on the assumption that protein complexes are densely connected sub-graphs in PPI networks. In this research, a dense sub-graph detection algorithm is first developed following this assumption by using clique seeds and graph entropy. Although the proposed algorithm generates a large number of reasonable predictions and its f-score is better than many previous algorithms, it still cannot identify many known protein complexes. After that, we analyze characteristics of known yeast protein complexes and find that not all of the complexes exhibit dense structures in PPI networks. Many of them have a star-like structure, which is a very special case of the core-attachment structure and it cannot be identified by many previous core-attachment-structure-based algorithms. To increase the prediction accuracy of protein complex identification, a multiple-topological-structure-based algorithm is proposed to identify protein complexes from PPI networks. Four single-topological-structure-based algorithms are first employed to detect raw predictions with clique, dense, core-attachment and star-like structures, respectively. A merging and trimming step is then adopted to generate final predictions based on topological information or GO annotations of predictions. A comprehensive review about the identification of protein complexes from static PPI networks to dynamic PPI networks is also given in this study. Secondly, genetic diseases often involve the dysfunction of multiple genes. Various types of evidence have shown that similar disease genes tend to lie close to one another in various biomolecular networks. The identification of disease genes via multiple data integration is indispensable towards the understanding of the genetic mechanisms of many genetic diseases. However, the number of known disease genes related to similar genetic diseases is often small. It is not easy to capture the intricate gene-disease associations from such a small number of known samples. Moreover, different kinds of biological data are heterogeneous and no widely acceptable criterion is available to standardize them to the same scale. In this study, a flexible and reliable multiple data integration algorithm is first proposed to identify disease genes based on the theory of Markov random fields (MRF) and the method of Bayesian analysis. A novel global-characteristic-based parameter estimation method and an improved Gibbs sampling strategy are introduced, such that the proposed algorithm has the capability to tune parameters of different data sources automatically. However, the Markovianity characteristic of the proposed algorithm means it only considers information of direct neighbors to formulate the relationship among genes, ignoring the contribution of indirect neighbors in biomolecular networks. To overcome this drawback, a kernel-based MRF algorithm is further proposed to take advantage of the global characteristics of biological data via graph kernels. The kernel-based MRF algorithm generates predictions better than many previous disease gene identification algorithms in terms of the area under the receiver operating characteristic curve (AUC score). However, it is very time-consuming, since the Gibbs sampling process of the algorithm has to maintain a long Markov chain for every single gene. Finally, to reduce the computational time of the MRF-based algorithm, a fast and high performance logistic-regression-based algorithm is developed for identifying disease genes from biomolecular networks. Numerical experiments show that the proposed algorithm outperforms many existing methods in terms of the AUC score and running time. To summarize, this study has developed several computational algorithms for identifying protein complexes and disease genes from biomolecular networks, respectively. These proposed algorithms are better than many other existing algorithms in the literature.

The major antinutritive factor in Leucaena for ruminants is condensed tannin (CT). CT bind proteins, incurring a negative effect on protein utilisation. The two major factors affecting the ability of CT to bind protein have been purported to be CT size and the pH of the reaction environment. To test these hypotheses the protein precipitating capacities of CT extracted from four promising Leucaena genotypes, L. leucocephala (K636), L. pallida (CQ3439), L. trichandra (CPI46568), and L. collinsii (OFI52/88) were assessed. L. leucocephala had approximately half the ability to precipitate protein on a g/g basis than L. pallida or L. trichandra while L. collinsii gave no measurable ability to precipitate protein (reaction environment=pH 5.0). Increasing or decreasing the pH of the reaction solution away from pH 5.0 (the isoelectric point of the protein) reduced the ability of CT from all the species to precipitate protein; the decrease being higher a pH 2.5 than at pH 7.5. At pH 2.5 L. leucocephala CT completely lost its capacity to precipitate protein. The relatively poor ability of L. leucocephala CT to bind protein at pH’s approximating those at the abomasum suggests L. leucocephala may have the greatest potential of the four Leucaena’s tested for increasing the extent of feed protein escaping ruminant degradation for later release and digestion in the small intestine, hence increasing the total amount of protein absorbed by ruminants. CT fractions from each Leucaena were also separated into individual CT’s, by size-exclusion chromatography and examined for protein precipitating capacity. In general it was found that the larger sized CT of the accessions L. pallida and L. trichandra could precipitate more protein than the smaller sized CT. This pattern was not found for L. leucocephala.

Thesis (Ph. D.)--University of Florida, 2004. / Typescript. Title from title page of source document. Document formatted into pages; contains 258 pages. Includes vita. Includes bibliographical references.

Le coenzyme Q est une molécule lipophile rédox rencontrée chez les eucaryotes et chez la plupart des procaryotes. La structure de Q correspond à une benzoquinone substituée par une chaîne polyisoprényle dont la longueur varie selon les organismes. Q joue le rôle de transporteur d'électrons dans les chaînes respiratoires d'où provient la plupart de l'énergie de la cellule. La biosynthèse de Q chez la bactérie Escherichia coli comporte huit étapes et implique au moins neuf protéines (UbiA-UbiH et UbiX). Trois réactions d'hydroxylation sont nécessaires pour la biosynthèse de Q8 en conditions aérobies. Alors que les protéines UbiH et UbiF présentent des homologies de séquence avec des monooxygénases à flavine connues pour catalyser des réactions d'hydroxylation, UbiB qui a été proposée comme étant la troisième hydroxylase, présente uniquement une homologie de séquence avec des kinases. Nous rapportons dans ce travail que la protéine VisC, renommée UbiI, catalyse la réaction d'hydroxylation auparavant attribuée à UbiB. Nous avons également identifié deux nouvelles protéines (YigP et YqiC, renommées respectivement UbiJ et UbiK) importantes pour le métabolisme de Q chez Escherichia coli puisque leur mutation diminue fortement le contenu en Q des souches mutantes. Ces protéines interagissent avec la plupart des protéines connues pour participer à la biosynthèse de Q ce qui implique l'existence d'un complexe de biosynthèse de Q. En utilisant des approches biochimiques et protéomiques, nous avons pu mettre en évidence un complexe impliquant plusieurs protéines Ubi et notamment UbiJ et UbiK. Ces deux protéines semblent avoir un rôle dans l'assemblage et/ou la stabilisation de ce complexe multiprotéique. Enfin, nous nous sommes intéressés à la biosynthèse de Q dans des conditions de cultures anaérobies. Nos résultats montrent l'existence « d'hydroxylases anaérobies », inconnues à ce jour, qui remplaçent les hydroxylases aérobies UbiH, UbiI et UbiF. Grâce à une approche phylogénétique, nous identifions un gène important pour la biosynthèse de Q uniquement en conditions anaérobies suggérant une réorganisation de la biosynthèse de Q entre ces deux environnements fréquemment rencontrés par E. coli. L'ensemble de nos résultats a permis d'améliorer notre connaissance de la voie de biosynthèse procaryote de Q grâce à la découverte de nouveaux gènes impliqués dans ce processus et grâce à l'identification de la fonction moléculaire de certaines protéines. / Ubiquinone (Q) is a lipophilic compound that plays an important role in electron and proton transport in the respiratory chains of Escherichia coli. Besides this important role in energy production, Q also functions as a membrane soluble antioxidant. The biosynthesis of Q8 requires eight reactions and involves at least nine proteins (UbiA-UbiH and UbiX) in Escherichia coli. Three of these reactions are hydroxylations resulting in the introduction of a hydroxyl group on carbon atoms at position 1, 5 and 6 of the aromatic ring. The C1 and C6 hydroxylation are well characterized whereas the C5 hydroxylation has been proposed to involve UbiB, a protein kinase without any sequence homology with monooxygenase. In this work, by genetic and biochemical methods we provide evidence that VisC which we renamed UbiI, displays sequence homology with monooxygenases and catalyzes the C5 hydroxylation, not UbiB. We have identified two new genes, yqiC and yigP (renammed UbiJ and UbiK) which are required only for Q8 biosynthesis in aerobic conditions. The exact role of the corresponding proteins, renamed UbiJ and UbiK, remains unknown. These proteins are able to interact with other Ubi proteins to be able to produce Q supporting the protein complex hypothesis. Our progress on the characterization of an Ubi-complex regrouping several Ubi proteins suggest that UbiJ and UbiK may fulfill functions related to the Ubi-complex stability. Mutants affected in hydroxylation steps are deficient for Q8 in aerobic conditions but recover a wild type Q8 content when grown in anaerobic conditions. This intriguing observation supports the existence of an alternative hydroxylation system independent from dioxygen which has not been characterized so far. By phylogenetic studies, we have identified a new gene in which the deletion affect the biosynthesis of Q only in anaerobic conditions suggesting a reorganization of Q biosynthesis in these two conditions. Our results has improved our knowledge of the prokaryotic Q biosynthetic pathway through the discovery of new genes involved in this process and through the identification of the molecular function of some proteins.

Le bourgeonnement est l'étape finale du cycle viral du virus VIH. Les particules virales vont devoir modifier la topologie de la membrane plasmique afin de promouvoir leur libération dans le milieu extracellulaire ; cette étape est réalisée par le recrutement de protéines ESCRT (en particulier CHMP4 et CHMP2) au point de bourgeonnement. A ce jour, les détails moléculaires de ce recrutement sont méconnus. Lethal Giant Discs (LgD) a été décrite dans la littérature comme un régulateur du traffic endosomal, et une interaction avec CHMP4B a été proposée pour l'orthologue humain CC2D1A. Un point majeur de ce travail aura été de caractériser l'interaction CC2D1A.CHMP4B, mais également de mieux comprendre l'organisation de la protéine. En particulier j'ai résolu la structure d'un fragment de LgD à 2.4 Å, comprenant une région hélicale et un domaine C2 en c-terminal. En outre, nous montrons que CC2D1A inhibe la capacité de CHMP4B à polymériser in vitro. A partir d'une structure cristallographique de CHMP4B et de données biochimiques, nous montrons que le site d'interaction de CC2D1A sur CHMP4B est impliqué dans la polymérisation de CHMP4B, et important pour la fonction de la protéine dans le contexte du bourgeonnement du HIV. Un projet parallèle m'a également conduit à définir un protocole de purification de la protéine CHMP2B recombinante sous forme monomérique, cet isoforme ayant été récemment impliqué dans la formation de structures tubulaires à la membrane plasmique et dans des activités de scission membranaire. En particulier, j'ai pû caractériser la protéine en présence de liposomes et préciser de nouveaux partenaires cellulaires. / Budding is the final step of HIV infection. Viral particles will have to modify the topology of the plasma membrane in order to achieve their correct release from the infected cell, by recruiting ESCRT proteins at the budding point, and among them CHMP4 and CHMP2 isoforms. So far, the molecular details of this recruitment are not precisely known.. Lethal Giant Discs (LgD) has been descibed in the litterature as a regulator of endosomal trafficking, and an interaction with CHMP4B has been proposed. A major point of this research is to propose a structural basis for this interaction, as well as a better understanding of the role and general organization of LgD/CC2D1A. The crystal structure of a LgD fragment (comprising a predicted coiled-coil motif and a c-terminal C2 domain) was solved in our lab at 2.4 A. Moreover, we show that CC2D1A impairs in vitro the ability of CHMP4B to polymerize. Based on a crystallographic structure of CHMP4B and biochemical data, we also show that the binding site of CC2D1A on CHMP4B is itself involved in polymerization, in the context of HIV budding. As a side project, I've also set up a protocole to obtain pure monomeric CHMP2B, which has been shown to polymerize at the plasma membrane, and I've characterized the protein in the presence of liposomes, along with new partners.

DNA double strand breaks are the most dangerous type of DNA damage. The MRN complex and 53BP1 have essential functions in the repair of DNA double strand breaks and are therefore important for maintaining genomic stability and preventing cancer. DNA double strand breaks are repaired by two main mechanisms - homologous recombination and non- homologous end joining. The MRN complex senses DNA double strand breaks and activates a cascade of posttranslational modifications that activates and recruits other effector proteins. In addition MRN mediated resection is important for removing adducts in non-homologous end joining and creating single stranded DNA required for homologous recombination. 53BP1 is recruited to DNA double strand breaks by site specific ubiquitinations and inhibits DNA resection, thereby promoting non-homologous end joining at the expense of homologous recombination. In this thesis we show that MRE11 binds to the R2TP chaperone complex through a CK2 mediated phosphorylation. Knockdown of R2TP or mutating the MRE11 binding site leads to decreased MRE11 levels and impaired DNA repair. Similar phenotype has been observed in cells from patients with ataxia-telangiectasia-like disorder (ATLD), containing MRE11 deletion mutation which is missing the R2TP complex binding site. Based on R2TP...

京都大学 / 0048 / 新制・課程博士 / 博士(情報学) / 甲第18405号 / 情博第520号 / 新制||情||92(附属図書館) / 31263 / 京都大学大学院情報学研究科知能情報学専攻 / (主査)教授 阿久津 達也, 教授 山本 章博, 教授 永持 仁 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

AlphaFold [Jumper et al., 2021, Evans et al., 2022] is a deep learning-based method that can accurately predict the structure of single- and multiple-chain proteins. However, its accuracy decreases with an increasing number of chains, and GPU memory limits the size of protein complexes that can be predicted. Recently, Elofsson’s groupintroduced a Monte Carlo tree search method, MoLPC, that can predict the structure of large complexes from predictions of sub-components [Bryant et al., 2022b]. However, MoLPC cannot adjust for errors in the sub-component predictions and requires knowledge of the correct protein stoichiometry. Large protein complexes are responsible for many essential cellular processes, such as mRNA splicing [Will and Lührmann, 2011], protein degradation [Tanaka, 2009], and protein folding [Ditzel et al., 1998]. However, the lack of structural knowledge of many large protein complexes remains challenging. Only a fraction of the eukaryoticcore complexes in CORUM [Giurgiu et al., 2019] have homologous structures covering all chains in PDB, indicating a significant gap in our structural understanding of protein complexes. AlphaFold-Multimer [Evans et al., 2022] is the only deep learning method that can predict the structure of more than two protein chains, trained on proteins of up to 20 chains, and can predict complexes of up to a few thousand residues, where memory limitations come into play. Another approach, MoLPC, is to predict the structure of sub-components of large complexes and assemble them. It has shown that it is possible to manually assemble large complexes from dimers manually [Burke et al., 2021] or use Monte Carlo tree search [Bryant et al., 2022b]. One limitation of the previous MoLPC approach is its inability to account for errors in sub-component prediction. The addition of small errors in each sub-component can propagate to a significant error when building the entire complex, leading toMoLPC’s failure. To overcome this challenge, the Monte Carlo Tree Search algorithms in MoLPC2 is enhanced to assemble protein complexes while simultaneously predicting their stoichiometry. Using MoLPC2, we accurately predicted the structures of 50 out of 175 non-redundant protein complexes (TM-score >0.8), while MoLPC only predicted 30. It should be noted that improvements introduced in AlphaFold version 2.3 enable the prediction of larger complexes, and if stoichiometry is known, it can accurately predict the structures of 74 complexes. Our findings suggest that assembling symmetrical complexes from sub-components results in higher accuracy while assembling asymmetrical complexes remains challenging.

No description available.

The purpose of this study was to completely characterize the protein moiety in the caroteno complex in C. poinsettae, determine if the distribution and level of protein in the pigment-protein complex in membranes of the wild type and in a colorless mutant could account for the differences in the stability of the membrane, and to determine if this protein is common to other pigmented and non-pigmented organisms. Also, electron microscopy of cell membranes of C. poinsettiae which had been exposed to gold-labelled antibody against the protein moitey of the pigment-protein complex, demonstrating that the protein is randomly distributed in the membranes of both wild type and colorless mutant.