Differential expression analysis for sequence count data via mixtures of negative binomials

The recent advent of Next-generation sequencing technologies has revolutionized the way of analyzing the genome. This innovation allows to get deeper information at a lower cost and in less time, and provides data that are discrete measurements. One of the most important applications with these data is the differential analysis, that is investigating if one gene exhibit a different expression level in correspondence of two (or more) biological conditions (such as disease states, treatments received and so on). As for the statistical analysis, the final aim will be statistical testing and for modeling these data the Negative Binomial distribution is considered the most adequate one especially because it allows for "over dispersion". However, the estimation of the dispersion parameter is a very delicate issue because few information are usually available for estimating it. Many strategies have been proposed, but they often result in procedures based on plug-in estimates, and in this thesis we show that this discrepancy between the estimation and the testing framework can lead to uncontrolled first-type errors. We propose a mixture model that allows each gene to share information with other genes that exhibit similar variability. Afterwards, three consistent statistical tests are developed for differential expression analysis. We show that the proposed method improves the sensitivity of detecting differentially expressed genes with respect to the common procedures, since it is the best one in reaching the nominal value for the first-type error, while keeping elevate power. The method is finally illustrated on prostate cancer RNA-seq data.

Identification of novel genetic alterations in pediatric cytogenetically normal acute myeloid leukemia by next-generation sequencing

Pediatric acute myeloid leukemia (AML) is a molecularly heterogeneous disease that arises from genetic alterations in pathways that regulate self-renewal and myeloid differentiation. While the majority of patients carry recurrent chromosomal translocations, almost 20% of childhood AML do not show any recognizable cytogenetic alteration and are defined as cytogenetically normal (CN)-AML. CN-AML patients have always showed a great variability in response to therapy and overall outcome, underlining the presence of unknown genetic changes, not detectable by conventional analyses, but relevant for pathogenesis, and outcome of AML. The development of novel genome-wide techniques such as next-generation sequencing, have tremendously improved our ability to interrogate the cancer genome. Based on this background, the aim of this research study was to investigate the mutational landscape of pediatric CN-AML patients negative for all the currently known somatic mutations reported in AML through whole-transcriptome sequencing (RNA-seq). RNA-seq performed on diagnostic leukemic blasts from 19 pediatric CN-AML cases revealed a considerable incidence of cryptic chromosomal rearrangements, with the identification of 21 putative fusion genes. Several of the fusion genes that were identified in this study are recurrent and might have a prognostic and/or therapeutic relevance. A paradigm of that is the CBFA2T3-GLIS2 fusion, which has been demonstrated to be a common alteration in pediatric CN-AML, predicting poor outcome. Important findings have been also obtained in the identification of novel therapeutic targets. On one side, the identification of NUP98-JARID1A fusion suggests the use of disulfiram; on the other, here we describe alteration-activating tyrosine kinases, providing functional data supporting the use of tyrosine kinase inhibitors to specifically inhibit leukemia cells. This study provides new insights in the knowledge of genetic alterations underlying pediatric AML, defines novel prognostic markers and putative therapeutic targets, and prospectively ensures a correct risk stratification and risk-adapted therapy also for the “all-neg” AML subgroup.

Data management and data analysis in the large European projects GEHA (GEnetics of Healthy Aging) and NU-AGE (NUtrition and AGEing): a bioinformatic approach

The aging process is characterized by the progressive fitness decline experienced at all the levels of physiological organization, from single molecules up to the whole organism. Studies confirmed inflammaging, a chronic low-level inflammation, as a deeply intertwined partner of the aging process, which may provide the “common soil” upon which age-related diseases develop and flourish. Thus, albeit inflammation per se represents a physiological process, it can rapidly become detrimental if it goes out of control causing an excess of local and systemic inflammatory response, a striking risk factor for the elderly population. Developing interventions to counteract the establishment of this state is thus a top priority. Diet, among other factors, represents a good candidate to regulate inflammation. Building on top of this consideration, the EU project NU-AGE is now trying to assess if a Mediterranean diet, fortified for the elderly population needs, may help in modulating inflammaging. To do so, NU-AGE enrolled a total of 1250 subjects, half of which followed a 1-year long diet, and characterized them by mean of the most advanced –omics and non –omics analyses. The aim of this thesis was the development of a solid data management pipeline able to efficiently cope with the results of these assays, which are now flowing inside a centralized database, ready to be used to test the most disparate scientific hypotheses. At the same time, the work hereby described encompasses the data analysis of the GEHA project, which was focused on identifying the genetic determinants of longevity, with a particular focus on developing and applying a method for detecting epistatic interactions in human mtDNA. Eventually, in an effort to propel the adoption of NGS technologies in everyday pipeline, we developed a NGS variant calling pipeline devoted to solve all the sequencing-related issues of the mtDNA.

Acinetobacter baumannii isolates from pets and horses in Switzerland: molecular characterization and clinical data

OBJECTIVES: We investigated whether Acinetobacter baumannii isolates of veterinary origin shared common molecular characteristics with those described in humans. METHODS: Nineteen A. baumannii isolates collected in pets and horses were analysed. Clonality was studied using repetitive extragenic palindromic PCR (rep-PCR) and multilocus sequence typing (MLST). PCR and DNA sequencing for various beta-lactamase, aminoglycoside-modifying enzyme, gyrA and parC, ISAba1 and IS1133, adeR and adeS of the AdeABC efflux pump, carO porin and class 1/2/3 integron genes were performed. RESULTS: Two main clones [A (n = 8) and B (n = 9)] were observed by rep-PCR. MLST indicated that clone A contained isolates of sequence type (ST) ST12 (international clone II) and clone B contained isolates of ST15 (international clone I). Two isolates of ST10 and ST20 were also noted. Seventeen isolates were resistant to gentamicin, 12 to ciprofloxacin and 3 to carbapenems. Isolates of ST12 carried bla(OXA-66), bla(ADC-25), bla(TEM-1), aacC2 and IS1133. Strains of ST15 possessed bla(OXA-69), bla(ADC-11), bla(TEM-1) and a class 1 integron carrying aacC1 and aadA1. ISAba1 was found upstream of bla(ADC) (one ST10 and one ST12) and/or bla(OXA-66) (seven ST12). Twelve isolates of different STs contained the substitutions Ser83Leu in GyrA and Ser80Leu or Glu84Lys in ParC. Significant disruptions of CarO porin and overexpressed efflux pumps were not observed. The majority of infections were hospital acquired and in animals with predisposing conditions for infection. CONCLUSIONS: STs and the molecular background of resistance observed in our collection have been frequently described in A. baumannii detected in human patients. Animals should be considered as a potential reservoir of multidrug-resistant A. baumannii.

Utilization of Probabilistic Models in Short Read Assembly from Second-Generation Sequencing

With the advent of cheaper and faster DNA sequencing technologies, assembly methods have greatly changed. Instead of outputting reads that are thousands of base pairs long, new sequencers parallelize the task by producing read lengths between 35 and 400 base pairs. Reconstructing an organism’s genome from these millions of reads is a computationally expensive task. Our algorithm solves this problem by organizing and indexing the reads using n-grams, which are short, fixed-length DNA sequences of length n. These n-grams are used to efficiently locate putative read joins, thereby eliminating the need to perform an exhaustive search over all possible read pairs. Our goal was develop a novel n-gram method for the assembly of genomes from next-generation sequencers. Specifically, a probabilistic, iterative approach was utilized to determine the most likely reads to join through development of a new metric that models the probability of any two arbitrary reads being joined together. Tests were run using simulated short read data based on randomly created genomes ranging in lengths from 10,000 to 100,000 nucleotides with 16 to 20x coverage. We were able to successfully re-assemble entire genomes up to 100,000 nucleotides in length.

Generation and analysis of a mouse intestinal metatranscriptome through Illumina based RNA-sequencing

With the advent of high through-put sequencing (HTS), the emerging science of metagenomics is transforming our understanding of the relationships of microbial communities with their environments. While metagenomics aims to catalogue the genes present in a sample through assessing which genes are actively expressed, metatranscriptomics can provide a mechanistic understanding of community inter-relationships. To achieve these goals, several challenges need to be addressed from sample preparation to sequence processing, statistical analysis and functional annotation. Here we use an inbred non-obese diabetic (NOD) mouse model in which germ-free animals were colonized with a defined mixture of eight commensal bacteria, to explore methods of RNA extraction and to develop a pipeline for the generation and analysis of metatranscriptomic data. Applying the Illumina HTS platform, we sequenced 12 NOD cecal samples prepared using multiple RNA-extraction protocols. The absence of a complete set of reference genomes necessitated a peptide-based search strategy. Up to 16% of sequence reads could be matched to a known bacterial gene. Phylogenetic analysis of the mapped ORFs revealed a distribution consistent with ribosomal RNA, the majority from Bacteroides or Clostridium species. To place these HTS data within a systems context, we mapped the relative abundance of corresponding Escherichia coli homologs onto metabolic and protein-protein interaction networks. These maps identified bacterial processes with components that were well-represented in the datasets. In summary this study highlights the potential of exploiting the economy of HTS platforms for metatranscriptomics.

REMOVING TECHNICAL VARIABILITY IN RNA-SEQ DATA USING CONDITIONAL QUANTILE NORMALIZATION

The ability to measure gene expression on a genome-wide scale is one of the most promising accomplishments in molecular biology. Microarrays, the technology that first permitted this, were riddled with problems due to unwanted sources of variability. Many of these problems are now mitigated, after a decade’s worth of statistical methodology development. The recently developed RNA sequencing (RNA-seq) technology has generated much excitement in part due to claims of reduced variability in comparison to microarrays. However, we show RNA-seq data demonstrates unwanted and obscuring variability similar to what was first observed in microarrays. In particular, we find GC-content has a strong sample specific effect on gene expression measurements that, if left uncorrected, leads to false positives in downstream results. We also report on commonly observed data distortions that demonstrate the need for data normalization. Here we describe statistical methodology that improves precision by 42% without loss of accuracy. Our resulting conditional quantile normalization (CQN) algorithm combines robust generalized regression to remove systematic bias introduced by deterministic features such as GC-content, and quantile normalization to correct for global distortions.

Streaming Support for Data Intensive Cloud-Based Sequence Analysis

Cloud computing provides a promising solution to the genomics data deluge problem resulting from the advent of next-generation sequencing (NGS) technology. Based on the concepts of “resources-on-demand” and “pay-as-you-go”, scientists with no or limited infrastructure can have access to scalable and cost-effective computational resources. However, the large size of NGS data causes a significant data transfer latency from the client’s site to the cloud, which presents a bottleneck for using cloud computing services. In this paper, we provide a streaming-based scheme to overcome this problem, where the NGS data is processed while being transferred to the cloud. Our scheme targets the wide class of NGS data analysis tasks, where the NGS sequences can be processed independently from one another. We also provide the elastream package that supports the use of this scheme with individual analysis programs or with workflow systems. Experiments presented in this paper show that our solution mitigates the effect of data transfer latency and saves both time and cost of computation.

Next-generation sequencing of HIV-1 RNA genomes: determination of error rates and minimizing artificial recombination.

Next-generation sequencing (NGS) is a valuable tool for the detection and quantification of HIV-1 variants in vivo. However, these technologies require detailed characterization and control of artificially induced errors to be applicable for accurate haplotype reconstruction. To investigate the occurrence of substitutions, insertions, and deletions at the individual steps of RT-PCR and NGS, 454 pyrosequencing was performed on amplified and non-amplified HIV-1 genomes. Artificial recombination was explored by mixing five different HIV-1 clonal strains (5-virus-mix) and applying different RT-PCR conditions followed by 454 pyrosequencing. Error rates ranged from 0.04-0.66% and were similar in amplified and non-amplified samples. Discrepancies were observed between forward and reverse reads, indicating that most errors were introduced during the pyrosequencing step. Using the 5-virus-mix, non-optimized, standard RT-PCR conditions introduced artificial recombinants in a fraction of at least 30% of the reads that subsequently led to an underestimation of true haplotype frequencies. We minimized the fraction of recombinants down to 0.9-2.6% by optimized, artifact-reducing RT-PCR conditions. This approach enabled correct haplotype reconstruction and frequency estimations consistent with reference data obtained by single genome amplification. RT-PCR conditions are crucial for correct frequency estimation and analysis of haplotypes in heterogeneous virus populations. We developed an RT-PCR procedure to generate NGS data useful for reliable haplotype reconstruction and quantification.

Tracking a tuberculosis outbreak over 21 years: strain-specific single nucleotide polymorphism-typing combined with targeted whole genome sequencing.

BACKGROUND  Whole genome sequencing (WGS) is increasingly used in molecular-epidemiological investigations of bacterial pathogens, despite cost- and time-intensive analyses. We combined strain-specific single nucleotide polymorphism (SNP)-typing and targeted WGS to investigate a tuberculosis cluster spanning 21 years in Bern, Switzerland. METHODS  Based on genome sequences of three historical outbreak Mycobacterium tuberculosis isolates, we developed a strain-specific SNP-typing assay to identify further cases. We screened 1,642 patient isolates, and performed WGS on all identified cluster isolates. We extracted SNPs to construct genomic networks. Clinical and social data were retrospectively collected. RESULTS  We identified 68 patients associated with the outbreak strain. Most were diagnosed in 1991-1995, but cases were observed until 2011. Two thirds belonged to the homeless and substance abuser milieu. Targeted WGS revealed 133 variable SNP positions among outbreak isolates. Genomic network analyses suggested a single origin of the outbreak, with subsequent division into three sub-clusters. Isolates from patients with confirmed epidemiological links differed by 0-11 SNPs. CONCLUSIONS  Strain-specific SNP-genotyping allowed rapid and inexpensive identification of M. tuberculosis outbreak isolates in a population-based strain collection. Subsequent targeted WGS provided detailed insights into transmission dynamics. This combined approach could be applied to track bacterial pathogens in real-time and at high resolution.

Highlights in Analytical Chemistry in Switzerland: OMA & OPA - A Software Tool for Mass Spectrometric Sequencing of Nucleic Acids

Oligonucleotides comprising unnatural building blocks, which interfere with the translation machinery, have gained increased attention for the treatment of gene-related diseases (e.g. antisense, RNAi). Due to structural modifications, synthetic oligonucleotides exhibit increased biostability and bioavailability upon administration. Consequently, classical enzyme-based sequencing methods are not applicable to their sequence elucidation and verification. Tandem mass spectrometry is the method of choice for performing such tasks, since gas-phase dissociation is not restricted to natural nucleic acids. However, tandem mass spectrometric analysis can generate product ion spectra of tremendous complexity, as the number of possible fragments grows rapidly with increasing sequence length. The fact that structural modifications affect the dissociation pathways greatly increases the variety of analytically valuable fragment ions. The gas-phase dissociation of oligonucleotides is characterized by the cleavage of one of the four bonds along the phosphodiester chain, by the accompanying loss of nucleases, and by the generation of internal fragments due to secondary backbone cleavage. For example, an 18-mer oligonucleotide yields a total number of 272’920 theoretical fragment ions. In contrast to the processing of peptide product ion spectra, which nowadays is highly automated, there is a lack of tools assisting the interpretation of oligonucleotide data. The existing web-based and stand-alone software applications are primarily designed for the sequence analysis of natural nucleic acids, but do not account for chemical modifications and adducts. Consequently, we developed a software to support the interpretation of mass spectrometric data of natural and modified nucleic acids and their adducts with chemotherapeutic agents.

Genome and transcriptome sequencing identifies breeding targets in the orphan crop tef (Eragrostis tef)

Background: Tef (Eragrostis tef), an indigenous cereal critical to food security in the Horn of Africa, is rich in minerals and protein, resistant to many biotic and abiotic stresses and safe for diabetics as well as sufferers of immune reactions to wheat gluten. We present the genome of tef, the first species in the grass subfamily Chloridoideae and the first allotetraploid assembled de novo. We sequenced the tef genome for marker-assisted breeding, to shed light on the molecular mechanisms conferring tef's desirable nutritional and agronomic properties, and to make its genome publicly available as a community resource. Results: The draft genome contains 672 Mbp representing 87% of the genome size estimated from flow cytometry. We also sequenced two transcriptomes, one from a normalized RNA library and another from unnormalized RNASeq data. The normalized RNA library revealed around 38000 transcripts that were then annotated by the SwissProt group. The CoGe comparative genomics platform was used to compare the tef genome to other genomes, notably sorghum. Scaffolds comprising approximately half of the genome size were ordered by syntenic alignment to sorghum producing tef pseudo-chromosomes, which were sorted into A and B genomes as well as compared to the genetic map of tef. The draft genome was used to identify novel SSR markers, investigate target genes for abiotic stress resistance studies, and understand the evolution of the prolamin family of proteins that are responsible for the immune response to gluten. Conclusions: It is highly plausible that breeding targets previously identified in other cereal crops will also be valuable breeding targets in tef. The draft genome and transcriptome will be of great use for identifying these targets for genetic improvement of this orphan crop that is vital for feeding 50 million people in the Horn of Africa.

A MOLECULAR APPROACH FOR DETECTING GENETIC DAMAGE USING THE ECO R1 FAMILY OF HUMAN DNA (SEQUENCING)

A clone of the primary Eco R1 family of human DNA sequences has been used as an indicator sequence for detecting alterations induced by a toxic agent. Specific clones of this family have been examined and compared to the consensus sequence to determine the normal variability of this family. Though variations were observed, data indicated that such clones can be used to study induced DNA modifications. This DNA was exposed to the toxic agent dimethyl sulfate under various conditions and a distinct pattern of aberrations was shown to occur. It is suggested that this approach be used to characterize patterns of damage induced by various agents in the ultimate development of a system capable of monitoring human genotoxic exposure. ^

RNA-SEQUENCING APPLICATIONS: GENE EXPRESSION QUANTIFICATION AND METHYLATOR PHENOTYPE IDENTIFICATION

My dissertation focuses on two aspects of RNA sequencing technology. The first is the methodology for modeling the overdispersion inherent in RNA-seq data for differential expression analysis. This aspect is addressed in three sections. The second aspect is the application of RNA-seq data to identify the CpG island methylator phenotype (CIMP) by integrating datasets of mRNA expression level and DNA methylation status. Section 1: The cost of DNA sequencing has reduced dramatically in the past decade. Consequently, genomic research increasingly depends on sequencing technology. However it remains elusive how the sequencing capacity influences the accuracy of mRNA expression measurement. We observe that accuracy improves along with the increasing sequencing depth. To model the overdispersion, we use the beta-binomial distribution with a new parameter indicating the dependency between overdispersion and sequencing depth. Our modified beta-binomial model performs better than the binomial or the pure beta-binomial model with a lower false discovery rate. Section 2: Although a number of methods have been proposed in order to accurately analyze differential RNA expression on the gene level, modeling on the base pair level is required. Here, we find that the overdispersion rate decreases as the sequencing depth increases on the base pair level. Also, we propose four models and compare them with each other. As expected, our beta binomial model with a dynamic overdispersion rate is shown to be superior. Section 3: We investigate biases in RNA-seq by exploring the measurement of the external control, spike-in RNA. This study is based on two datasets with spike-in controls obtained from a recent study. We observe an undiscovered bias in the measurement of the spike-in transcripts that arises from the influence of the sample transcripts in RNA-seq. Also, we find that this influence is related to the local sequence of the random hexamer that is used in priming. We suggest a model of the inequality between samples and to correct this type of bias. Section 4: The expression of a gene can be turned off when its promoter is highly methylated. Several studies have reported that a clear threshold effect exists in gene silencing that is mediated by DNA methylation. It is reasonable to assume the thresholds are specific for each gene. It is also intriguing to investigate genes that are largely controlled by DNA methylation. These genes are called “L-shaped” genes. We develop a method to determine the DNA methylation threshold and identify a new CIMP of BRCA. In conclusion, we provide a detailed understanding of the relationship between the overdispersion rate and sequencing depth. And we reveal a new bias in RNA-seq and provide a detailed understanding of the relationship between this new bias and the local sequence. Also we develop a powerful method to dichotomize methylation status and consequently we identify a new CIMP of breast cancer with a distinct classification of molecular characteristics and clinical features.

Diversity of zooplankton assemblages at Station L4, Western English Channel, measured using next generation DNA sequencing

Background: Zooplankton play an important role in our oceans, in biogeochemical cycling and providing a food source for commercially important fish larvae. However, difficulties in correctly identifying zooplankton hinder our understanding of their roles in marine ecosystem functioning, and can prevent detection of long term changes in their community structure. The advent of massively parallel Next Generation Sequencing technology allows DNA sequence data to be recovered directly from whole community samples. Here we assess the ability of such sequencing to quantify the richness and diversity of a mixed zooplankton assemblage from a productive monitoring site in the Western English Channel. Methodology/Principle Findings: Plankton WP2 replicate net hauls (200 µm) were taken at the Western Channel Observatory long-term monitoring station L4 in September 2010 and January 2011. These samples were analysed by microscopy and metagenetic analysis of the 18S nuclear small subunit ribosomal RNA gene using the 454 pyrosequencing platform. Following quality control a total of 419,042 sequences were obtained for all samples. The sequences clustered in to 205 operational taxonomic units using a 97% similarity cut-off. Allocation of taxonomy by comparison with the National Centre for Biotechnology Information database identified 138 OTUs to species level, 11 to genus level and 1 to order, <2.5% of sequences were classified as unknowns. By comparison a skilled microscopic analyst was able to routinely enumerate only 75 taxonomic groups. Conclusions: The percentage of OTUs assigned to major eukaryotic taxonomic groups broadly aligns between the metagenetic and morphological analysis and are dominated by Copepoda. However, the metagenetics reveals a previously hidden taxonomic richness, especially for Copepoda and meroplankton such as Bivalvia, Gastropoda and Polychaeta. It also reveals rare species and parasites. We conclude that Next Generation Sequencing of 18S amplicons is a powerful tool for estimating diversity and species richness of zooplankton communities.