Refine
Has Fulltext
- yes (1054)
Year of publication
- 2022 (1054) (remove)
Document Type
- Journal article (810)
- Doctoral Thesis (199)
- Working Paper (17)
- Book article / Book chapter (12)
- Preprint (4)
- Report (3)
- Book (2)
- Master Thesis (2)
- Review (2)
- Bachelor Thesis (1)
Language
- English (1054) (remove)
Keywords
- COVID-19 (16)
- Datennetz (14)
- SARS-CoV-2 (11)
- machine learning (11)
- virtual reality (11)
- heart failure (10)
- prostate cancer (10)
- inflammation (9)
- remote sensing (9)
- 3D printing (8)
Institute
- Theodor-Boveri-Institut für Biowissenschaften (130)
- Graduate School of Life Sciences (79)
- Institut für Informatik (66)
- Medizinische Klinik und Poliklinik I (56)
- Neurologische Klinik und Poliklinik (54)
- Medizinische Klinik und Poliklinik II (50)
- Institut für Geographie und Geologie (48)
- Klinik und Poliklinik für Anästhesiologie (ab 2004) (44)
- Institut für Organische Chemie (41)
- Institut für Psychologie (39)
Schriftenreihe
Sonstige beteiligte Institutionen
- IZKF Nachwuchsgruppe Geweberegeneration für muskuloskelettale Erkrankungen (5)
- Bernhard-Heine-Centrum für Bewegungsforschung (4)
- Zentraleinheit Klinische Massenspektrometrie (3)
- Fraunhofer-Institut für Silicatforschung ISC (2)
- Helmholtz Institute for RNA-based Infection Research (HIRI) (2)
- Krankenhaushygiene und Antimicrobial Stewardship (Universitätsklinikum) (2)
- Wilhelm-Conrad-Röntgen-Forschungszentrum für komplexe Materialsysteme (2)
- ALPARC - The Alpine Network of Protected Areas (1)
- Albert-Ludwigs-Universität Freiburg (1)
- Arizona State University, Tempe, Arizona, USA (1)
Distributed practice is a well-known learning strategy whose beneficial effects on long-term learning are well proven by various experiments. In learning from texts, the benefits of distribution might even go beyond distributed practice, i.e. distribution of repeated materials. In realistic learning scenarios as for example school or university learning, the reader might read multiple texts that not repeat but complement each other. Therefore, distribution might also be implemented between multiple texts and benefit long-term learning in analogy to distributed practice. The assumption of beneficial effects of this distributed learning can be deduced from theories about text comprehension as the landscape model of reading (van den Broek et al., 1996) in combination with theories of desirable difficulties in general (R. A. Bjork & Bjork, 1992) and distributed practice in particular (Benjamin & Tullis, 2010). This dissertation aims to investigate (1) whether distributed learning benefits learning; (2) whether the amount of domain-specific prior knowledge moderates the effects of distribution, (3) whether distributed learning affects the learner’s meta-cognitive judgments in analogy to distributed practice and (4) whether distributed practice is beneficial for seventh graders in learning from single text.
In Experiment 1, seventh graders read two complementary texts either massed or distributed by a lag of one week between the texts. Learning outcomes were measured immediately after reading the second text and one week later. Judgements of learning were assessed immediately after each text. Experiment 2 replicated the paradigm of Experiment 1 while shortening the lag between the texts in the distributed condition to 15 min. In both experiments, an interaction effect between learning condition (distributed vs. massed) and retention interval (immediate vs. delayed) was found. In the distributed condition, the participants showed no decrease in performance between the two tests, whereas participants in the massed condition did. However, no beneficial effects were found in the delayed test for the distributed condition but even detrimental effects for the distributed condition in the immediate test. In Experiment 1, participants in the distributed condition perceived learning as less difficult but predicted lower success than the participants in the massed condition.
Experiment 3 replicated the paradigm of Experiment 1 with university students in the laboratory. In the preregistered Experiment 4, an additional retention interval of two weeks was realized. In both experiments, the same interaction between learning condition and retention interval was found. In Experiment 3, the participants in the distributed condition again showed no decrease in performance between the two tests, whereas participants in the massed condition did. However, even at the longer retention interval in Experiment 4, no beneficial effects were found for the distributed condition. Domain-specific prior knowledge was positively associated with test performance in both experiments. In Experiment 4, the participants with low prior knowledge seemed to be impaired by distributed learning, whereas no difference was found for participants with medium or high prior knowledge.
In the preregistered Experiment 5, seventh graders read a single text twice. The rereading took place either massed or distributed with one week. Immediately after rereading, judgements of learning were assessed. Learning outcomes were assessed four min after second reading or one week later. Participants in the distributed condition predicted lower learning success than participants in the massed condition. An interaction effect between learning condition and retention interval was found, but no advantage for the distributed condition. Participants with low domain-specific prior knowledge showed lower performance in short-answer questions in the distributed condition than in the massed condition.
Overall, the results seem less encouraging regarding the effectiveness of distribution on learning from single and multiple texts. However, the experiments reported here can be perceived as first step in the realistic investigation of distribution in learning from texts.
Protein-DNA interactions are central to many biological processes and form the bedrock of gene transcription, DNA replication, and DNA repair processes. Many proteins recognize specific sequences in DNA- a restriction enzyme must only cut at the correct sequence and a transcription factor should bind at its consensus sequence. Some proteins are designed to bind to specific structural or chemical features in DNA, such as DNA repair proteins and some DNA modifying enzymes. Target-specific DNA binding proteins initially bind to non-specific DNA and then search for their target sites through different types of diffusion mechanisms. Atomic force microscopy (AFM) is a single-molecule technique that is specifically well-suited to resolve the distinct states of target-specific as well as nonspecific protein-DNA interactions that are vital for a deeper insight into the target site search mechanisms of these enzymes. In this thesis, protein systems involved in epigenetic regulation, base excision repair (BER), and transcription are investigated by single-molecule AFM analyses complemented by biochemical and biophysical experiments.
The first chapter of this thesis narrates the establishment of a novel, user-unbiased MatLab-based tool for automated DNA bend angle measurements on AFM data. This tool has then been employed to study the initial lesion detection step of several DNA glycosylases. These results promoted a model describing the altered plasticities of DNA at the target lesions of DNA glycosylases as the fundamental mechanism for their enhanced efficiency of lesion detection.
In the second chapter of this thesis, the novel automated tool has been further extended to provide protein binding positions on the DNA along with corresponding DNA bend angles and applied to the study of DNMT3A DNA methyltransferase. These AFM studies revealed preferential co-methylation at specific, defined distances between two CpG sites by the enzyme and when combined with biochemical analyses and structural modelling supported novel modes of CpG co-methylation by DNMT3A.
In the third chapter of this thesis, the role of 8-oxo-guanine glycosylase (hOGG1) in Myc-mediated transcription initiation has been investigated. AFM analyses revealed that in the presence of oxidative damage in DNA, Myc is recruited to its target site (E-box) by hOGG1 through direct protein-protein interactions, specifically under oxidizing conditions. Intriguingly, oxidation of hOGG1 was further observed to result in dimerization of hOGG1, which may also play a role in the mechanism of transcription regulation by hOGG1 under oxidative stress.
Detecting anomalies in transaction data is an important task with a high potential to avoid financial loss due to irregularities deliberately or inadvertently carried out, such as credit card fraud, occupational fraud in companies or ordering and accounting errors. With ongoing digitization of our world, data-driven approaches, including machine learning, can draw benefit from data with less manual effort and feature engineering. A large variety of machine learning-based anomaly detection methods approach this by learning a precise model of normality from which anomalies can be distinguished. Modeling normality in transactional data, however, requires to capture distributions and dependencies within the data precisely with special attention to numerical dependencies such as quantities, prices or amounts.
To implicitly model numerical dependencies, Neural Arithmetic Logic Units have been proposed as neural architecture. In practice, however, these have stability and precision issues.
Therefore, we first develop an improved neural network architecture, iNALU, which is designed to better model numerical dependencies as found in transaction data. We compare this architecture to the previous approach and show in several experiments of varying complexity that our novel architecture provides better precision and stability.
We integrate this architecture into two generative neural network models adapted for transaction data and investigate how well normal behavior is modeled. We show that both architectures can successfully model normal transaction data, with our neural architecture improving generative performance for one model.
Since categorical and numerical variables are common in transaction data, but many machine learning methods only process numerical representations, we explore different representation learning techniques to transform categorical transaction data into dense numerical vectors. We extend this approach by proposing an outlier-aware discretization, thus incorporating numerical attributes into the computation of categorical embeddings, and investigate latent spaces, as well as quantitative performance for anomaly detection.
Next, we evaluate different scenarios for anomaly detection on transaction data. We extend our iNALU architecture to a neural layer that can model both numerical and non-numerical dependencies and evaluate it in a supervised and one-class setting. We investigate the stability and generalizability of our approach and show that it outperforms a variety of models in the balanced supervised setting and performs comparably in the one-class setting. Finally, we evaluate three approaches to using a generative model as an anomaly detector and compare the anomaly detection performance.
Although the field of fungal infections advanced tremendously, diagnosis of invasive pulmonary aspergillosis (IPA) in immunocompromised patients continues to be a challenge. Since IPA is a multifactorial disease, investigation from different aspects may provide new insights, helpful for improving IPA diagnosis. This work aimed to characterize the human immune response to Aspergillus fumigatus in a multilevel manner to identify characteristic molecular candidates and risk factors indicating IPA, which may in the future support already established diagnostic assays. We combined in vitro studies using myeloid cells infected with A. fumigatus and longitudinal case-control studies investigating patients post allogeneic stem cell transplantation (alloSCT) suffering from IPA and their match controls.
Characteristic miRNA and mRNA signatures indicating A. fumigatus-infected monocyte-derived dendritic cells (moDCs) demonstrated the potential to differentiate between A. fumigatus and Escherichia coli infection. Transcriptome and protein profiling of alloSCT patients suffering from IPA and their matched controls revealed a distinctive IPA signature consisting of MMP1 induction and LGAL2 repression in combination with elevated IL-8 and caspase-3 levels. Both, in vitro and case-control studies, suggested cytokines, matrix-metallopeptidases and galectins are important in the immune response to A. fumigatus. Identified IPA characteristic molecular candidates are involved in numerous processes, thus a combination of these in a distinctive signature may increase the specificity. Finally, low monocyte counts, severe GvHD of the gut (grade ≥ 2) and etanercept administration were significantly associated with IPA diagnosis post alloSCT. Etanercept in monocyte-derived macrophages (MDM) infected with A. fumigatus downregulates genes involved in the NF-κB and TNF-α pathway and affects the secretion of CXCL10.
Taken together, identified characteristic molecular signatures and risk factors indicating IPA may in the future in combination with established fungal biomarkers overcome current diagnostic challenges and help to establish tailored antifungal therapy. Therefore, further multicentre studies are encouraged to evaluate reported findings.
Abstract
Purpose: To compare ab interno trabeculectomy by trabecular meshwork excision to plasma-mediated ablation in primary open-angle glaucoma.
Methods: Retrospectively collected data of TrabEx+ (n=56) and Trabectome (n=99) were compared by coarsened exact matching to reduce confounding and matched based on baseline intraocular pressure and age. Primary outcomes were intraocular pressure and the number of glaucoma medications. Complications and the need for additional glaucoma surgery were assessed. Patients were followed for up to one year.
Results: 53 TrabEx+ could be matched to Trabectome. Baseline intraocular pressure was 16.5 ± 4.6 mmHg in both; age was 73.7 ± 8.8 years and 71.5 ± 9.9 years in TrabEx+ and Trabectome, respectively. TrabEx+ were taking more medications than Trabectome (p<0.001). Intraocular pressure was reduced to 14.8±4.3 in TrabEx+ and 13.4±3.4 in Trabectome at 6 months, and to 14.9±6.0 (p=0.13) in TrabEx+ and to 14.1±3.8 mmHg (all p<0.05) in Trabectome at 12 months. Medications were reduced at both 6 and 12 months (p< 0.05). No differences were seen between both groups at 6 and 12 months. In TrabEx+, only one serious complication occurred, and two patients required further glaucoma surgery.
Conclusion: Although both groups had a baseline intraocular pressure considered low for ab interno trabeculectomy, intraocular pressure and medications were reduced further at 6 and 12 months. Intraocular pressure reduction did not reach significance in TrabEx+ at 12 months. The inter-group comparison did not reveal any significant differences. Both had a low complication rate.
Microbial, mammalian and plant cells produce and contain secondary metabolites, which typically are soluble in water to prevent cell damage by crystallization. The formation of ion pairs, e.g. with carboxylic acids or mineral acids, is a natural blueprint to keep basic metabolites in solution. It was aimed at showing whether the mostly large carboxylates form soluble protic ionic liquids (PILs) with basic natural products resulting in enhanced aqueous solubility. Furthermore, their supramolecular pattern in aqueous solution was studied. Thereby, naturally occurring carboxylic acids were identified being appropriate counterions for natural basic compounds and facilitate the formation of PILs with their beneficial characteristics, like improved dissolution rate and enhanced apparent solubility.
In the field of biofabrication, biopolymer-based hydrogels are often used as bulk materials with defined structures or as bioinks. Despite their excellent biocompatibility, biopolymers need chemical modification to fulfill mechanical stability.
In this thesis, the primary alcohol of hyaluronic acid was oxidized using TEMPO/TCC oxidation to generate aldehyde groups without ring-opening mechanism of glycol cleavage using sodium periodate. For crosslinking reaction of the aldehyde groups, adipic acid dihydrazide was used as bivalent crosslinker for Schiff Base chemistry. This hydrogel system with fast and reversible crosslinking mechanism was used successfully as bulk hydrogel for chondrogenic differentiation with human mesenchymal stem cells (hMSC).
Gelatin was modified with pentenoic acid for crosslinking reaction via light controllable thiol-ene reaction, using thiolated 4-arm sPEG as multivalent crosslinker. Due to preservation of the thermo responsive property of gelatin by avoiding chain degradation during modification reaction, this gelatin-based hydrogel system was successfully processed via 3D printing with low polymer concentration. Good cell viability was achieved using hMSC in various concentrations after 3D bioprinting and chondrogenic differentiation showed promising results.
Context. In active galaxies, matter is accreted onto super massive black holes (SMBH). This accretion process causes a region roughly the size of our solar system to outshine the entire host galaxy, forming an active galactic nucleus (AGN). In some of these active galaxies, highly relativistic particle jets are formed parallel to the rotation axis of the super massive black hole. A fraction of these sources is observed under a small inclination angle between the pointing direction of the jet and the observing line of sight. These sources are called blazars. Due to the small inclination angle and the highly relativistic speeds of the particles in the jet, beaming effects occur in the radiation of these particles. Blazars can be subdivided into the high luminosity flat spectrum radio quasars (FSRQs) and the low luminosity BL Lacertae objects (BL Lacs). As all AGN, blazars are broadband emitters and therefore observable from the longest wavelengths in the radio regime to the shortest wavelengths in the gamma-ray regime. In this thesis I will analyze blazars at these two extremes with respect to their parsec-scale properties in the radio and their time evolution properties in gamma-ray flux.
Method. In the radio regime the technique of very long baseline interferometry (VLBI) can be used in order to spatially resolve the synchrotron radiation coming from those objects down to sub-parsec scales. This information can be used to observe the time evolution of the structure of such sources. This is done in large monitoring programs such as the MOJAVE (15 GHz) and the Boston University blazar monitoring program (43 GHz). In this thesis I utilize data of 28 sources from these monitoring programs spanning 10 years of observation from 2003 to 2013, resulting in over 1800 observed epochs, to study the brightness temperature and diameter gradients of these jets. I conduct a search for systematic geometry transitions in the radio jets. The synchrotron cooling time in the radio core of the jets is used to determine the magnetic field strength in the radio core. Considering the jet geometry, these magnetic field strengths are scaled to the ergosphere of the SMBH in order to obtain the distance of the radio core to the SMBH.
In the gamma-regime these blazars cannot be spatially resolved. Due to this, it is hard to put strong constrains onto where the gamma-ray emitting region is. Blazars have shown to be variable at high energies on time scales down to minutes. The nature of this variability can be studied in order to put constrains on the particle acceleration mechanism and possibly the region and size of the gamma-ray emitting region. The variability of blazars in the energy range between 0.1 GeV and 1 GeV can for example be observed with the pair-conversion telescope on board the Fermi satellite. I use 10 years of data from the Fermi-LAT (LAT: Large Area Telescope) satellite in order to study the variability of a large sample of blazars (300-800, depending on the used significance filters for data points). I quantify this variability with the Ornstein-Uhlenbeck (OU) parameters and the power spectral density (PSD) slopes. The same procedure is applied to 20 light curves available for the radio sample.
Results. The diameter evolution along the jet axis of the radio sources suggests, that FSRQs feature flatter gradients than BL Lacs. Fitting these gradients, it is revealed that BL Lacs are systematically better described by a simple single power law than FSRQs. I found 9 sources with a strongly constrained geometry transition. The sources are 0219+421, 0336-019, 0415+379, 0528+134, 0836+710, 1101+384, 1156+295, 1253-055 and 2200+420. In all of these sources, the geometry transition regions are further out in the jet than the Bondi sphere. The magnetic field strengths of BL Lacs is systematically larger than that of FSRQs. However the scaling of these fields suggest that the radio cores of BL Lac objects are closer to the SMBHs than the radio cores of FSRQs. Analyzing the variability of Fermi-LAT light curves yields consistent results for all samples. FSRQs show systematically steeper PSD slopes and feature OU parameters more favorable to strong variability than BL Lacs. The Fermi-LAT light curves of the sub-sample of radio jets, suggest an anticorrelation between the jet complexity from the radio observations and the OU-parameters as well as the PSD slopes from the gamma-ray observations.
Conclusion.
The flatter diameter gradients of FSRQs suggest that these sources are more collimated further down the jet than BL Lacs. The systematically better description of the diameter and brightness temperature gradient by a single power law of BL Lacs, suggest that FSRQs are more complex with respect to the diameter evolution along the jet and the surface brightness distribution than BL Lac objects. FSRQs often feature regions where recollimation can occur in distinct knots within the jets. For the sources where a geometry transition could be constrained, the Bondi radius, being systematically smaller than the position of the transition region along the jet axis, suggest that changing pressure gradients are not the sole cause for these systematic geometry transitions. Nevertheless they may be responsible for recollimation regions, found typically downstream the jet, beyond the Bondi radius and the transition zone. The difference in the distance of the radio cores between FSRQs and BL Lacs is most likely due to the combination of differences in SMBH masses and systematically smaller jet powers in BL Lacs. The variability in energy ranges above 100 MeV and above 1 GeV-regime suggest that many light curves of BL Lac objects are more likely to be white noise while the PSD slopes and the OU parameters of FSRQ gamma-ray light curves favor stronger variability on larger time scales with respect to the time binning of the analyzed light curve. Although the anticorrelation of the jet complexity acquired from the radio observations and the PSD slopes and OU parameters from the gamma-observations suggest that more complex sources favor OU parameters and PSD slopes resulting in more variability (not white noise) it is beyond the scope of this thesis to pinpoint whether this correlation results from causation. The question whether a complex jet causes more gamma-ray variability or more gamma-ray variability causes more complex jets cannot be answered at this point. Nevertheless the computed correlation measures suggest that this dependence is most likely not linear and therefore an indication that these effects might even interact.
Parkinson’s Disease (PD) constitutes a major healthcare burden in Europe. Accounting for aging alone, ~700,000 PD cases are predicted by 2040. This represents an approximately 56% increase in the PD population between 2005 and 2040, with a consequent rise in annual disease‐related medical costs. Gait and balance disorders are a major problem for patients with PD and their caregivers, mainly because to their correlation with falls. Falls occur as a result of a complex interaction of risk factors. Among them, Freezing of Gait (FoG) is a peculiar gait derangement characterized by a sudden and episodic inability to produce effective stepping, causing falls, mobility restrictions, poor quality of life, and increased morbidity and mortality. Between 50–70% of PD patients have FoG and/or falls after a disease duration of 10 years, only partially and inconsistently improved by dopaminergic treatment and Deep Brain Stimulation (DBS). Treatment-induced worsening has been also observed under certain conditions. Effective treatments for gait disturbances in PD are lacking, probably because of the still poor understanding of the supraspinal locomotor network.
In my thesis, I wanted to expand our knowledge of the supraspinal locomotor network and in particular the contribution of the basal ganglia to the control of locomotion. I believe this is a key step towards new preventive and personalized therapies for postural and gait problems in patients with PD and related disorders. In addition to patients with PD, my studies also included people affected by Progressive Supranuclear Palsy (PSP). PSP is a rare primary progressive parkinsonism characterized at a very early disease stage by poor balance control and frequent backwards falls, thus providing an in vivo model of dysfunctional locomotor control.
I focused my attention on one of the most common motor transitions in daily living, the initiation of gait (GI). GI is an interesting motor task and a relevant paradigm to address balance and gait impairments in patients with movement disorders, as it is associated with FoG and high risk of falls. It combines a preparatory (i.e., the Anticipatory Postural Adjustments [APA]) and execution phase (the stepping) and allows the study of movement scaling and timing as an expression of muscular synergies, which follow precise and online feedback information processing and integration into established feedforward patterns of motor control.
By applying a multimodal approach that combines biomechanical assessments and neuroimaging investigations, my work unveiled the fundamental contribution of striatal dopamine to GI in patients with PD. Results in patients with PSP further supported the fundamental role of the striatum in GI execution, revealing correlations between the metabolic intake of the left caudate nucleus with diverse GI measurements. This study also unveiled the interplay of additional brain areas in the motor control of GI, namely the Thalamus, the Supplementary Motor Area (SMA), and the Cingulate cortex. Involvement of cortical areas was also suggested by the analysis of GI in patients with PD and FoG. Indeed, I found major alterations in the preparatory phase of GI in these patients, possibly resulting from FoG-related deficits of the SMA. Alterations of the weight shifting preceding the stepping phase were also particularly important in PD patients with FoG, thus suggesting specific difficulties in the integration of somatosensory information at a cortical level. Of note, all patients with PD showed preserved movement timing of GI, possibly suggesting preserved and compensatory activity of the cerebellum. Postural abnormalities (i.e., increased trunk and thigh flexion) showed no relationship with GI, ruling out an adaptation of the motor pattern to the altered postural condition. In a group of PD patients implanted with DBS, I further explored the pathophysiological functioning of the locomotor network by analysing the timely activity of the Subthalamic Nucleus (STN) during static and dynamic balance control (i.e., standing and walking). For this study, I used novel DBS devices capable of delivering stimulation and simultaneously recording Local Field Potentials (LFP) of the implanted nucleus months and years after surgery. I showed a gait-related frequency shift in the STN activity of PD patients, possibly conveying cortical (feedforward) and cerebellar (feedback) information to mesencephalic locomotor areas. Based on this result, I identified for each patient a Maximally Informative Frequency (MIF) whose power changes can reliably classify standing and walking conditions. The MIF is a promising input signal for new DBS devices that can monitor LFP power modulations to timely adjust the stimulation delivery based on the ongoing motor task (e.g., gait) performed by the patient (adaptive DBS).
Altogether my achievements allowed to define the role of different cortical and subcortical brain areas in locomotor control, paving the way for a better understanding of the pathophysiological dynamics of the supraspinal locomotor network and the development of tailored therapies for gait disturbances and falls prevention in PD and related disorders.
Clostridioides difficile is a bacterial species well known for its ability to cause C. difficile
infection (also known as CDI). The investigation of the role of this species in the human
gut has been so far dominated by a disease-centred perspective, focused on studying
C. difficile in relation to its associated disease.
In this context, the first aim of this thesis was to combine publicly available
metagenomic data to analyse the microbial composition of stool samples from patients
diagnosed with CDI, with a particular focus on identifying a CDI-specific microbial
signature.
However, similarly to many other bacterial species inhabiting the human gut, C.
difficile association with disease is not valid in absolute terms, as C. difficile can be
found also among healthy subjects. Further aims of this thesis were to 1) identify
potential C. difficile reservoirs by screening a wide range of habitats, hosts, body sites
and age groups, and characterize the biotic context associated with C. difficile
presence, and 2) investigate C. difficile within-species diversity and its toxigenic
potential across different age groups.
The first part of the thesis starts with the description of the concepts and
definitions used to identify bacterial species and within-species diversity, and then
proceeds to provide an overview of the bacterial species at the centre of my
investigation, C. difficile. The first Chapter includes a detailed description of the
discovery, biology and physiology of this clinically relevant species, followed by an
overview of the diagnostic protocols used in the clinical setting to diagnose CDI.
The second part of the thesis describes the methodology used to investigate
the questions mentioned above, while the third part presents the results of such
investigative effort. I first show that C. difficile could be found in only a fraction of the
CDI samples and that simultaneous colonization of multiple enteropathogenic species
able to cause CDI-like clinical manifestations is more common than previously
thought, raising concerns about CDI overdiagnosis. I then show that the CDIassociated
gut microbiome is characterized by a specific microbial signature,
distinguishable from the community composition associated with non-CDI diarrhea.
Beyond the nosocomial and CDI context, I show that while rarely found in adults, C.
difficile is a common member of the infant gut microbiome, where its presence is
associated with multiple indicators typical of a desirable healthy microbiome
development.
In addition, I describe C. difficile extensive carriage among asymptomatic
subjects, of all age groups and a potentially novel clade of C. difficile identified
exclusively among infants.
Finally, I discuss the limitations, challenges and future perspectives of my
investigation.