The peroxisome is a subcellular organelle that carries out a diverse range of metabolic functions, including the b-oxidation of very long chain fatty acids, the breakdown of peroxide and the a-oxidation of fatty acids. Disruption of peroxisome metabolic functions leads to severe disease in humans. These diseases can be broadly grouped into two categories: those in which a single enzyme is defective, and those known as the peroxisome biogenesis disorders (PBDs), which result from a generalised failure to import peroxisomal matrix proteins (and consequently result in disruption of multiple metabolic pathways). The PBDs result from mutations in PEX genes, which encode protein products called peroxins, required for the normal biogenesis of the peroxisome. PEX1 encodes an AAA ATPase that is essential for peroxisome biogenesis, and mutations in PEX1 are the most common cause of PBDs worldwide. This study focused on the identification of mutations in PEX1 in an Australasian cohort of PBD patients, and the impact of these mutations on PEX1 function. As a result of the studies presented in this thesis, twelve mutations in PEX1 were identified in the Australasian cohort of patients. The identified mutations can be broadly grouped into three categories: missense mutations, mutations directly introducing a premature termination codon (PTC) and mutations that interrupt the reading frame of PEX1. The missense mutations that were identified were R798G, G843D, I989T and R998Q; all of these mutations affect amino acid residues located in the AAA domains of the PEX1 protein. Two mutations that directly introduce PTCs into the PEX1 transcript (R790X and R998X), and four frameshift mutations (A302fs, I370fs, I700fs and S797fs) were identified. There was also one mutation found in an intronic region (IVS22-19A>G) that is presumed to affect splicing of the PEX1 mRNA. Three of these mutations, G843D, I700fs and G973fs, were found at high frequency in this patient cohort. At the commencement of these studies, it was hypothesised that missense mutations would result in attenuation of PEX1 function, but mutations that introduced PTCs, either directly or indirectly, would have a deleterious effect on PEX1 function. Mutations introducing PTCs are thought to cause mRNA to be degraded by the nonsense-mediated decay of mRNA (NMD) pathway, and thus result in a decrease in PEX1 protein levels. The studies on the cellular impact of the identified PEX1 mutations were consistent with these hypotheses. Missense mutations were found to reduce peroxisomal protein import and PEX1 protein levels, but a residual level of function remained. PTC-generating mutations were found to have a major impact on PEX1 function, with PEX1 mRNA and protein levels being drastically reduced, and peroxisomal protein import capability abolished. Patients with two missense mutations showed the least impact on PEX1 function, patients with two PTC-generating mutations had a severe defect in PEX1 function, and patients carrying a combination of a missense mutation and a PTC-generating mutation showed levels of PEX1 function that were intermediate between these extremes. Thus, a correlation between PEX1 genotype and phenotype was defined for the Australasian cohort of patients investigated in these studies. For a number of patients, mutations in the coding sequence of one PEX1 allele could not be identified. Analysis of the 5' UTR of this gene was therefore pursued for potential novel mutations. The initial analyses demonstrated that the 5' end of PEX1 extended further than previously reported. Two co-segregating polymorphisms were also identified, termed –137 T>C and –53C>G. The -137T>C polymorphism resided in an upstream, in-frame ATG (termed ATG1), and the possibility that the additional sequence represented PEX1 coding sequence was examined. While both ATGs were found to be functional by virtue of in vitro and in vivo expression investigations, Western blot analysis of the PEX1 protein in patient and control cell extracts indicated that physiological translation of PEX1 was from the second ATG only. Using a luciferase reporter approach, the additional sequence was found to exhibit promoter activity. When examined alone the -137T>C polymorphism exerted a detrimental effect on PEX1 promoter activity, reducing activity to half that of wild-type levels, and the -53C>G polymorphism increased PEX1 promoter activity by 25%. When co-expressed (mimicking the physiological condition) these polymorphisms compensated for each other to bring PEX1 promoter activity to near wild-type levels. The PEX1 mutations identified in this study have been utilised by collaborators at the National Referral Laboratory for Lysosomal, Peroxisomal and Related Genetic Disorders (based at the Women's and Children's Hospital, Adelaide), in prenatal diagnosis of the PBDs. In addition, the identification of three common mutations in Australasian PBD patients has led to the implementation of screening for these mutations in newly referred patients, often enabling a precise diagnosis of a PBD to be made. Finally, the strong correlation between genotype and phenotype for the patient cohort investigated as part of these studies has generated a basis for the assessment of newly identified mutations in PEX1.

Master of Science / Department of Diagnostic Medicine/Pathobiology / Roman Reddy Ganta / Ehrlichia chaffeensis a tick transmitted Anaplasmataceae family pathogen responsible for human monocytic ehrlichiosis. Differential gene expression appears to be an important pathogen adaptation mechanism for its survival in dual hosts. One of the ways to test this hypothesis is by performing mutational analysis that aids in altering the expression of genes. Mutagenesis is also a useful tool to study the effects of a gene function in an organism. Focus of my research has been to prepare several modified Himar transposon mutagenesis constructs for their value in introducing mutations in E. chaffeensis genome. While the work is in progress, research team from our group used existing Himar transposon mutagenesis plasmids and was able to create mutations in E. chaffeensis. Multiple mutations were identified by Southern blot analysis. I redirected my research efforts towards mapping the genomic insertion sites by performing the semi-random two step PCR (ST-PCR) method, followed by DNA sequence analysis. In this method, the first PCR is performed with genomic DNA as the template with a primer specific to the insertion segment and the second primer containing an anchored degenerate sequence segment. The product from the first PCR is used in the second PCR with nested transposon insertion primer and a primer designed to bind to the known sequence portion of degenerate primer segment. This method aided in identifying the genomic locations of four E. chaffeensis mutants and also was valuable in confirming four other sites mapped previously by the rescue cloning method. This is the first mutational analysis study in the genome of an Ehrlichia species. Mapping the genomic transposon insertion sites is the first critical step needed for the continued research to define the importance of the mutations in understanding the pathogenesis caused by the organism.

Objectives: Increasing knowledge of genetics has found that a mutation to the BRCA 1 or 2 genes are associated with a high risk of developing breast cancer throughout the lifespan. A woman with this genetic mutation may consider preventive surgery to reduce the risk of breast cancer. This involves a prophylactic bilateral mastectomy to remove the breasts when there is no cancer present and may be followed by breast reconstruction. This study aimed to explore the lived experience and psycho-social impact on women of this surgery. Design: Interpretative phenomenological analysis was employed in an in-depth study of a small sample of eleven female patients with BRCA 1/2 genetic mutations who had undergone preventive surgery of prophylactic bilateral mastectomy. Methods: Semi-structured interviews were carried out. The transcripts of those interviews served as the data for an interpretative phenomenological analysis. Results and conclusions: Three themes were identified from the Interpretative Phenomenological Analysis to convey the lived experience of participants. These were (1) focus on reduced risk of cancer; taking control, relief and benefit finding, (2) a focus on relationships; family life, medical professional and BRCA support group and other women with lived experience, and (3) Focus on experiencing surgery and impact on self; the importance of reconstruction, loss of sexual attractiveness, impact on self from negative reaction of others and adjusting to surgical results. The implications are discussed in relation to the current literature and clinical practice.

Οικογενής Υπερχοληστερολαιμία (FH) είναι η επικρατής αυτοσωμική νόσος, κατά την οποία τα επίπεδα χοληστερόλης στο αίμα είναι αυξημένα, εμφανίζονται ξανθώματα και ένα αυτοσωμικό επικρατές χαρακτηριστικό για στεφανιαία αρτηριακή νόσος (CAD). Η FH προκαλείται από ανωμαλία στο γονίδιο LDL-R και κάποιες φορές και στο γονίδιο APOB (apolipoprotein B-100). Η ετεροζυγία του LDLR συναντάται σε αναλογία πληθυσμού 1:500. Πρόσφατα παρατηρήθηκε ότι και το γονίδιο PCSK9 (proprotein convertase subtilisin/kexin type 9) προκαλεί FH. Τα γονίδια APOB και PCSK9 αποκλείστηκαν από τη συγκεκριμένη έρευνα. Στόχοι της διατριβής ήταν (α) η καταγραφή των μεταλλάξεων του γονιδίου LDLR (Low Density Lipoprotein Receptor) σε 21 πληθυσμούς, (β) ο υπολογισμός της συχνότητας αυτών των μεταλλάξεων και (γ) η προσθήκη αυτών των δεδομένων σε μία γενετική βάση δεδομένων, την FINDbase, η οποία δίνει πληροφορίες για τη συχνότητα μιας μετάλλαξης σε κάθε χώρα καθώς και το φαρμακευτικό δείκτη της. Από τους 21 πληθυσμούς, οι 14 προέρχονταν από Ευρωπαϊκές χώρες (Ελλάδα, Γερμανία,Πορτογαλία, Τσεχία, Ολλανδία, Ισπανία, Βρετανία, Ιταλία, Πολωνία, Σουηδία, Γαλλία, Αυστρία, Βέλγιο και Δανία) και οι υπόλοιποι από την Κίνα, την Ιαπωνία, την Μαλαισία, το Λίβανο, τις Φιλιππίνες, την Ταϊβάν και το Καναδά. Τα δεδομένα των μεταλλάξεων σε κάθε πληθυσμό αντλήθηκαν από άρθρα (papers) μέσω της Βάσης Δεδομένων Pubmed και της μηχανής αναζήτησης Google. Τα άρθρα επιλέχθηκαν με βάση (1) το μέγεθος του δείγματος και (2) τη χρονολογία πραγματοποίησης της έρευνας στο συγκεκριμένο πληθυσμό. Η συχνότητα υπολογίστηκε σε σύνολο χρωμοσωμάτων, δηλαδή στο διπλάσιο του μεγέθους του δείγματος. Ως ιδανικό μέγεθος δείγματος θεωρήθηκε ένα σύνολο τουλάχιστον 100 χρωμοσωμάτων, δηλ. 50 άτομα. Η καταγραφή των δεδομένων έγινε σε λογιστικό φύλλο Excel. Τα αποτελέσματα έδειξαν ότι υπάρχει μεγάλη ανομοιογένεια σε επίπεδο μεταλλάξεων ανάμεσα στους 21 πληθυσμούς. / Familial hypercholesterolaemia (FH), defined as the heritable occurence of severe hypercholesterolaemia with cholesterol deposits in tendons and premature heart disease, is caused by at least four genes in sterol and lipoprotein pathways and displays varying gene-dose effects. The genes are the low-density lipoprotein (LDL) receptor, apolipoprotein (apo) B, proprotein convertase subtilisin/kexin 9, and the autosomal recessive hypercholesterolaemia (ARH) adaptor protein. The world-wide prevalence of FH is about 1 in 500 people. In this assessment, the genes apoB, PCSK9 and ARH have been excluded. The aim of this study was the recording of LDLR mutations in 21 populations, the calculation of the mutations’ frequencies in each population and the introduction of these data in the National Ethnic Mutation DataBase (NEΜDB), FINDbase, which gives information about a mutation’s frequency in each country and also about its pharmacogenomic marker. Among 21 populations, 14 were of European origin (Greece, Germany, Portugal, Czech Republic, Netherlands, Spain, Great Britain, Italy, Poland, Sweden, France, Austria, Belgium and Denmark) and the remainders from China, Japan, Malaysia, Lebanon, Philippines, Taiwan and Canada. The mutation data in each population were derived from papers through the database of references, PubMed and the search engine, Google. The selection of papers was based on (1) the size of patient group and (2) the date of paper publication. The calculation of mutation frequency was based on the total number of chromosomes, which was the double size of the patient group. An ideal size of sample was at least 100 chromosomes, which means 50 index patients. The data were inserted in an excel file. The results showed that there is a great ανομοιογένεια in mutation level among 21 populations.

Domain swapping is a wide spread phenomenon which involves the association between two or more protein subunits such that intra-molecular interactions between domains in each subunit are replaced by equivalent inter-molecular interactions between the same domains in different subunits. This thesis is devoted to the analysis of the factors that drive proteins to undergo such association modes. The specific system analyzed is the monomer to swapped dimer formation of the B1 domain of the immunoglobulin G binding protein (GB1). The formation of this dimer was shown to be fostered by 4 amino acid substitutions (L5V, F30V, Y33F, A34F) (Byeon et al., 2003). In this work, computational protein design and molecular dynamics simulations, both with detailed atomic models, were used to gain insight into how these 4 mutations may promote the domain swapping reaction. The stability of the wt and quadruple mutant GB1 monomers was assessed using the software DESIGNER, a fully automatic procedure that selects amino acid sequences likely to stabilize a given backbone structure (Wernisch et al., 2000). Results suggest that 3 of the mutations (L5V, F30V, A34F) have a destabilizing effect. The first mutation (L5V) forms destabilizing interactions with surrounding residues, while the second (F30V) is engaged in unfavorable interactions with the protein backbone, consequently causing local strain. Although the A34F substitution itself is found to contribute favorably to the stability of the monomer, this is achieved only at the expense of forcing the wild type W43 into a highly strained conformation concomitant with the formation of unfavorable interactions with both W43 and V54. Finally, we also provide evidence that A34F mutation stabilizes the swapped dimer structure. Although we were unable to perform detailed protein design calculations on the dimer, due to the lower accuracy of the model, inspection of its 3D structure reveals that the 34F side chains pack against one another in the core of the swapped structure, thereby forming extensive non-native interactions that have no counterparts in the individual monomers. Their replacement by the much smaller Ala residue is suggested to be significantly destabilizing by creating a large internal cavity, a phenomenon, well known to be destabilizing in other proteins. Our analysis hence proposes that the A34F mutation plays a dual role, that of destabilizing the GB1 monomer structure while stabilizing the swapped dimer conformation. In addition to the above study, molecular dynamics simulations of the wild type and modeled quadruple mutant GB1 structures were carried out at room and elevated temperatures (450 K) in order to sample the conformational landscape of the protein near its native monomeric state, and to characterize the deformations that occur during early unfolding. This part of the study was aimed at investigating the influence of the amino acid sequence on the conformational properties of the GB1 monomer and the possible link between these properties and the swapping process. Analysis of the room temperature simulations indicates that the mutant GB1 monomer fluctuates more than its wild type counter part. In addition, we find that the C-terminal beta-hairpin is pushed away from the remainder of the structure, in agreement with the fact that this hairpin is the structural element that is exchanged upon domain swapping. The simulations at 450 K reveal that the mutant protein unfolds more readily than the wt, in agreement with its decreased stability. Also, among the regions that unfold early is the alpha-helix C-terminus, where 2 out of the 4 mutations reside. NMR experiments by our collaborators have shown this region to display increased flexibility in the monomeric state of the quadruple mutant. Our atomic scale investigation has thus provided insights into how sequence modifications can foster domain swapping of GB1. Our findings indicate that the role of the amino acid substitutions is to decrease the stability of individual monomers while at the same time increase the stability of the swapped dimer, through the formation of non-native interactions. Both roles cooperate to foster swapping.

La variation quantitative résulte de l'action combinée des gènes et de leur environnement. Pour comprendre la relation génotype-phénotype et disséquer l'architecture des caractères complexes, deux approches sont couramment employées. D'une part l'évolution expérimentale qui permet de quantifier le nombre et l'effet des mutations dans la construction d'un phénotype soumis à une pression de sélection, d'autre part la cartographie de QTL (Quantitative Trait Loci) et/ou la génétique d'association qui permettent d'identifier les locus responsables de la variation phénotypique. Au cours de cette thèse, nous avons combiné l'ensemble de ces approches pour (1) évaluer le rôle relatif des nouvelles mutations et de la variabilité résiduelle dans la réponse à la sélection ; (2) identifier les déterminants génétiques sous tendant cette réponse ; (3) disséquer, pour un locus candidat, les mécanismes génétiques de sa contribution à la variation phénotypique. Pour cela, nous disposons d'un matériel génétique résultant d'une expérience de sélection divergente pour la date de floraison menée depuis plus de dix ans. Cette expérience a été conduite en parallèle à partir de deux lots de semences de lignées commerciales de maïs (F252 et MBS847). Pour chaque lignée de départ, deux populations ont été constituées, une population précoce et une population tardive produites en sélectionnant et autofécondant les génotypes les plus précoces/tardifs à chaque génération. Nous avons caractérisé la réponse à la sélection après 7 générations. Cette réponse est rapide, asymétrique entre populations et significative dans 3 des 4 populations. Elle est linéaire avec le temps ce qui indique que des nouvelles mutations contribuent à créer de la variance génétique à chaque génération. Nous avons identifié un locus majeur contribuant à 35% de la variation pour la date de floraison dans la population F252 tardive et pour lequel les deux allèles étaient présents dans le lot de semence initial sous forme d'hétérozygotie résiduelle. Les deux allèles présentent des haplotypes très divergents autant au niveau de leur variation nucléotidique (5.7%) que d'un point de vue structural (16 indels) sur une région proche du gène eIF-4A (Eukaryotic Initiation Translation Factor 4A). L'association de ce locus avec la date de floraison et d'autres caractères corrélés tels que la hauteur et le nombre de feuilles a été confirmée par une caractérisation développementale fine de génotypes précoces et tardifs et également dans un panel d'association comprenant 317 lignées de maïs cultivé. En plus d'un effet pléiotrope, nous avons montré grâce au développement de méthodes statistiques que ce locus présente des interactions épistatique fortes avec d'autres locus en ségrégation puisque son effet dépend largement du fond génétique. Nous avons finalement utilisé des AFLP (Amplified Fragment Length Polymorphisms) sur tous les génotypes issus des 7 premières générations de sélection afin d'identifier d'autres polymorphismes potentiellement impliqués dans la réponse à la sélection. Nos résultats préliminaires montrent une différenciation génétique et épigénétique entre les populations sélectionnées qui semble être préférentiellement due à de l'hétérozygotie résiduelle.

Mitochondrial DNA encodes thirteen protein subunits in the oxidative phosphorylation system (OXPHOS) that is responsible for cellular energy production. Mitochondrial disorders have been identified to be associated with mtDNA mutations. However, the molecular mechanisms of specific mtDNA mutations are still being explored in order to establish causative links. This study tries to elucidate the mutational effects of mtDNA on OXPHOS complex activities and cell growths. Using mouse 3T3 fibroblasts as a cell model, single-cell clones with different growth rates were isolated. The entire mtDNA genome was sequenced for mutations. The enzymatic activities of OXPHOS complex I to V were analysed. Three growth patterns represented by five clones were identified. Three clones (clone #2, #3, and #6) had the shortest doubling times (11.5 - 14.9 hours). Clone #1 had a medium growth rate (19.2 hous); and clone #5 had a significantly slow growth rate (22 hours). MtDNA sequencing results revealed that clone #5 had several heteroplasmic mutations (one in 16S rRNA, two in tRNAser (UCN), three in tRNAasp, one in tRNAlys, one in COI, five in COII, and one in ATPase8) while the other four clones showed sequence homology. Enzymatic analyses showed that on average clone #5 had significantly low complex III, IV, and V activities (p < 0.05). Changes in biochemical properties and protein structure were analyzed to deduct possible mechanisms for reduced respiration. In conclusion, the slow growth rate is associated with reduced OXPHOS enzyme functions. It is most likely that the combination of COI and COII mutations resulted in the reduction of complex IV function. It is still unclear whether the ATPase8 mutation (T7869A) in the non-conserved region alone can have such a pronounced phenotypic effect. A reduction in complex III also cannot be explained since there were no mutations in the only mtDNA-encoded complex III gene, but it is possible that there are mutations in the nDNA-encoded complex III genes. Mutations in tRNA and rRNA genes may also be responsible for reduced protein syntheses and consequently reduced OXPHOS activities. It is unclear why complex I activity was not affected. Although the mutational effect of individual mtDNA mutation observed cannot be clearly identified, this study establishes a correlation between mtDNA mutation and cell energy production and growth.

Stress can be defined as factors which reduce an individual’s survival and/or reproductive ability. Certain stressors strike harder against individuals the more harmful mutations they carry, thus increasing selection against harmful mutations. The aim of this project is to test if exposure to stress, during many generations, decreases the frequency of harmful mutations and lead to more adapted individuals, or if specific adaptations to the stressor override this effect and results in reduced adaptation in the original environment through trade-offs. To test these hypotheses, I use selection lines of Drosophila melanogaster, where the flies during the larval state either have been exposed to media with reduced nutritional value or a heat shock treatment over 22 generations. The results from this experiment show that the selection lines have adapted to their novel environments, since survival of the larvae had increased in the novel environment they had been exposed to for 22 generations. However, survival of selection lines were if anything decreased rather than elevated in the original environment. A plausible explanation to these results is that adaptations to stressors occur both through trade-offs and reduced frequency of generally harmful mutations, but that the effect of the former possibly is larger than the latter. / Stress kan definieras som faktorer som minskar en individs överlevnad och/eller reproduktiva förmåga. Vissa stressmiljöer slår relativt hårdare mot individer ju fler skadliga mutationer de bär, och ökar därför selektionstrycket mot skadliga mutationer. Detta projekt har som mål att testa om exponering mot sådana stressmiljöer, under flera generationer, minskar frekvensen av skadliga mutationer och leder till generellt bättre anpassade individer, eller om specifika anpassningar till stressmiljön överskuggar denna effekt och via trade-offs leder till individer som är sämre anpassade till ursprungsmiljön. För att testa dessa hypoteser använder jag mig här av selektionslinjer av Drosophila melanogaster, där flugorna under larvstadiet antingen utsatts för en näringsfattig miljö eller en värmechock under 22 generationer. Resultaten från detta experiment visar en tydlig evolutionär respons, i och med att larvöverlevnaden ökat för de selekterade linjerna i den stressmiljö de utsatts för efter 22 generationer. Test av överlevnad i ursprungsmiljön visar dock ingen signifikant skillnad mot kontrollinjerna, men om något att de selekterade linjerna klarade sig något sämre. Dessa resultat tyder om något på att anpassningar som skett till den nya miljön på bekostnad av anpassningar i ursprungsmiljön (via trade-offs) överskuggar ökad anpassning via en minskad frekvens av generellt skadliga mutationer.

Bei der vorliegenden Arbeit handelt es sich um eine experimentelle Untersuchung zu Unterschieden in der Signaltransduktion von drei mutierten Thyrotropin-Rezeptoren (mTSHR), die in Rattenschilddrüsenzellen stabil exprimiert werden. Mittels Proteomanalyse konnte die Proteinexpression von mehreren Signalproteinen quantitativ untersucht werden und mit der Proteinexpression in Zellen, die Wildtyp (WT)-Rezeptoren exprimieren, verglichen werden. In den Zellen mit mutierten Rezeptoren konnte eine gesteigerte Expression von Proteinen nachgewiesen werden, die in der Endozytose eine Schlüsselrolle spielen. Weiterhin konnte hier eine verstärkte Internalisierung der mTSHR festgestellt werden. Der Phosphatidyl-Inositol-3-Kinase (PI3K) Signalweg wurde in mTSHR-Schilddrüsenzellen im Vergleich zu WT-Zellen verstärkt aktiviert, und dies geschah Proteinkinase A (PKA) -unabhängig. Zudem konnte eine vorübergehende Aktivierung des Proteins Exchange protein directly activated by cAMP (Epac) durch das zyklische Adenosinmonophosphat (cAMP) nachgewiesen werden, die möglicherweise mit der Aktivierung des PI3K-Signalwegs in Zusammenhang steht. Die Ergebnisse gewähren somit neue Einblicke in die biologische Aktivität von mTSHR. Zum einen bleiben wahrscheinlich die mTSHR nach Internalisierung weiterhin aktiv, zum anderen reduziert sich die weitere Signaltransduktion nicht nur auf cAMP-PKA/Inositoltrisphosphat-Signalwege. Diese Ergebnisse könnten zu einem besseren Verständnis für den Verlauf der Schilddrüsenautonomie beitragen.

Millions of somatic mutations in human cancers have been identified by sequenc- ing. Identifying and distinguishing cancer driver genes amongst the millions of candi- date mutations remains a major challenge. Accurate identification of driver genes and mutations is essential for the progress of cancer research and personalizing treatment based on accurate stratification of patients. Because of inter-tumor genetic hetero- geneity, numerous driver mutations within a gene can be found at low frequencies. This makes them difficult to differentiate from other non-driver mutations. Inspired by these challenges, we devised a novel way of identifying cancer driver genes. Our approach utilizes multiple complementary types of information, specifically cellular phenotypes, cellular locations, function, and whole body physiological phenotypes as features. We demonstrate that our method can accurately identify known cancer driver genes and distinguish between their role in different types of cancer. In ad- dition to identifying known driver genes, we identify several novel candidate driver genes. We provide an external evaluation of the predicted genes using a dataset of 26 nasopharyngeal cancer samples that underwent whole exome sequencing. We find that the predicted driver genes have a significantly higher rate of mutation than non-driver genes, both in publicly available data and in the nasopharyngeal cancer samples we use for validation. Additionally, we characterize sub-networks of genes that are jointly involved in specific tumors.