Decades of research in quantitative genetics have found substantial heritability for personality traits as well as for mental disorders which formed the basis of the ongoing molecular genetic studies that aim to identify genetic variations that actually contribute to the manifestation of complex traits. With regard to psychological traits, genetic variation impacting neurotransmitter function have been of particular interest. Additionally, the role of environmental factors including gene × environment interactions has been further investigated and the impor-tance of developmental aspects has been stressed. Furthermore, endophenotypes which link complex traits with their respective biological underpinnings and thus bridge the gap between gene and behaviour have begun to be included in research efforts. In accordance with this approach, this thesis aims to further examine the influence of genetic variation impacting serotonergic and dopaminergic functioning on endophenotypes of anxiety-related behaviour. To this end, two well established paradigms – the acoustic startle reflex and the cortisol stress response – were employed. Both show considerable interindividual variation which has been found in quantitative genetic studies to be at least partly based on genetic factors. In addition, the neural circuits underlying these endophenotypes are relatively well understood and thus reveal references for the detection of associated genetic influences. The results of this thesis associate the overall startle magnitude in two independent samples of young adults with a polymorphism in the promoter region of the serotonin transporter (5-HTT) gene (5-HTTLPR): Carriers of the short (S) allele which results in a reduced gene ex-pression showed a stronger startle magnitude which is in line with numerous findings linking the S allele to increased measures of negative emotionality. In addition to 5-HTTLPR, the effects of past stressful life events on the startle response were investigated: Participants who had recently experienced at least one stressful life event exhibited stronger startle responses and reduced habituation of the startle reflex although there was no 5-HTTLPR × environment inter-action effect. A third study revealed independent and joint effects of 5-HTTLPR and a poly-morphism in the dopamine receptor 4 gene (DRD4) in the same sample of young adults with regard to the cortisol stress response with carriers of the DRD4 7R allele which has been associ-ated with higher scores in sensation seeking, showing reduced cortisol responses. In addition, a 5-HTTLPR × DRD4 interaction effect emerged: 5-HTTLPR long (L) allele carriers showed the lowest cortisol response but only when they possessed at least one copy of the DRD4 7R allele. Moreover, in a fourth study a life span approach was taken and the influence of a further important serotonergic polymorphism which impacts the functioning of tryptophan hydroxylase 2 (TPH2), the rate limiting enzyme in the biosynthesis of serotonin, on interindividual differences in the startle response was investigated in three different age samples: children, young adults and older adults. There was a sex × TPH2 genotype interaction effect in a sample of young adults on the overall startle response while there was no effect of TPH2 in children or older adults. The last study of this thesis presents findings regarding the influence of two dopaminergic polymorphisms in genes encoding the enzyme catechol-O-methyltransferase (COMT) and the dopamine transporter (DAT), respectively, which both terminate dopamine signalling and are thus important regulators of dopaminergic neurotransmission, on the startle reflex in older adults. COMT met/met homozygotes showed the strongest and val/val homozygotes displayed the smallest startle magnitude which is in line with findings linking the COMT met allele to increased scores of anxiety related traits and disorders. Regarding DAT, participants homozygous for the 10R allele, which had previously associated with attention-deficit hyperactivity disorder, showed a stronger overall startle response. In sum, this thesis comprises data on interindividual differences in an electrophysiological and a hormonal endophenotype across the life span and their association with serotonergic and dopaminergic function based on genetic variation. One major finding is the clear evidence for the influence of serotonergic polymorphisms on the startle response in young adults while in contrast in older adults genetic variation in the dopaminergic system exerted considerable influence. These differences might be due to developmental processes in the different stages of life although cohort effects and effects of different recruitment strategies can also not be ruled out. Furthermore, there were significant differences regarding the genetic influence on the acoustic startle reflex and cortisol stress response in one and the same sample which might be due to methodological differences of the two paradigms as well as differences in their underlying neuronal circuits. In conclusion, this thesis supports the acoustic startle reflex and the cortisol stress response as valuable endophenotypes and thus indicators for underlying neurobiological circuits although some methodological issues remain. It also highlights the importance of taking developmental factors and changes over the course of life into account. Finally, this thesis emphasizes the necessity to include reliably and validly assessed past experienced events in molecular genetic studies in order to understand the interplay between genetic and environmental factors in shaping (endo)-phenotypes.

The 5-HTT gene (SLC6A4) is regulated by a common polymorphism in the promoter region (5-HTTLPR), which has functional consequences. Two major alleles have been observed and shown to have differential transcriptional activity with the long (L) allele having greater gene expression than the short (S) allele. 5-HTTLPR appears to modulate depression, anxiety and personality traits such as neuroticism. Additionally, a significant influence of 5-HTTLPR genotype on amygdala reactivity in response to fearful stimuli has been reported. Moreover, 5-HTTLPR seems to impact on the role of stressful life events (SLEs) in the development of depression. An elevated risk of depression and suicidal behaviors has been found in carriers of at least one low expressing S allele who had experienced SLEs, suggesting a gene x environment interaction. However, a recent meta-analysis showed that several findings failed to replicate this finding. Since genetic polymorphisms of the dopaminergic and serotonergic neurotransmission interact at the molecular, analyses with another polymorphism of the dopaminergic system, the dopamine D4 receptor (DRD4) was included to consider these likely gene-gene interactions (epistasis). The aim of this series of studies was to investigate the role 5-HTTLPR and SLEs on the endocrine stress response in different age samples. While newborns have been examined by a heel prick, stress responses were provoked in children (8-12 yrs) and younger adults (19-31 yrs) and older adults (54-68 yrs.) with the Trier Social Stress Test (TSST). The Life History Calendar (LHC) and Life Events Questionnaire (LEQ) were used to acquire data on SLEs. While in newborns the S/S genotype showed a significantly higher acute endocrine stress response than L/L or S/L genotypes, no significant difference between genotype groups was found in children. In the younger adult sample, the genotype impacted on cortisol stress responsiveness was reversed. Adults carrying the more active L allele of the 5-HTTLPR polymorphism showed a significantly larger cortisol response to the TSST than individuals carrying at least one of the lower expressing S allele. In older adults, no significant difference between genotype groups was found. However, results point in the same direction with showing highest cortisol response in individuals with L/L genotype. These data suggest that the association between 5-HTTLPR and endocrine stress reactivity seems to alter across lifespan, more specific the effects of genotype turns around. In addition, a significant interaction effect of 5-HTTLPR and SLEs has been found in the sample of younger adults, i.e. that early SLE as well as a severe number SLEs across the entire lifespan seem to modulate the interaction between HPA axis activity and 5-HTTLPR genotype. Additionally, a DRD4 by 5-HTTLPR interaction emerged which point to independent and joint effects of these polymorphisms on stress responsivity with regard to the concept of genegene interaction.

No description available.

With the progress of deep learning methods the last couple of years, object detection related tasks are improving rapidly. Using object detection for detecting guns in schools remove the need for human supervision and hopefully reduces police response time. This paper investigates how a gun detection system can be built by reading frames locally and using a server for detection. The detector is based on a pre-trained SSD model and through transfer learning is taught to recognize guns. The detector obtained an Average Precision of 51.1% and the server response time for a frame of size 1920 x 1080 was 480 ms, but could be scaled down to 240 x 135 to reach 210 ms, without affecting the accuracy. A non-gun class was implemented to reduce the number of false positives and on a set of 300 images containing 165 guns, the number of false positives dropped from 21 to 11.

(EN) GH/IGF-1 axis components (GH, growth hormone receptor (GH-R), IGF-1, IGF-1 receptor (IGF-1R), IGF-binding proteins (IGFBPs)) participate in the control of glucose metabolism, inflammatory processes as well as cell proliferation and differentiation, including adipocytes and monocytes. The aim of the present study was to evaluate the role of local mRNA expression of GH/IGF-1 axis components in subcutaneous adipose tissue (SCAT) and peripheral monocytes (PM) in the development of insulin resistance and differences of adipose tissue mass in following groups of patients: obese females with and without type 2 diabetes mellitus and subjects with active untreated acromegaly. A total number of 66 subjects were included in the study: obese females without type 2 diabetes mellitus (OB), obese females with type 2 diabetes mellitus (T2DM), acromegalic patients (AC) and healthy lean control subjects (C). T2DM underwent 2 weeks of very-low- calorie diet (VLCD - energy content 2500 kJ/day). According to our results we suggest that decreased mRNA expression of IGF-1, IGF-1R, IGFBP-2 and IGFBP-3 in adipose tissue of T2DM subjects may contribute to changes of fat differentiation capacity and the increased IGF-1R mRNA expression in peripheral monocytes in these patients may play a role in the regulation of...

La sécurité d’attachement joue un rôle significatif dans la gestion du stress. En effet, elle permet à l’enfant de se réguler, tant sur le plan comportemental, émotionnel que physiologique. Le présent mémoire, qui s’inscrit dans une optique de valorisation des données de recherche, vise à investiguer, d’un point de vue longitudinal, le lien entre la sécurité d’attachement et la synchronisation physiologique des mécanismes de stress chez les dyades mère-enfant. En étudiant la réactivité de l’axe hypothalamo-pituito-surrénalien (HPS) qui régule le cortisol, souvent désigné comme l’hormone du stress, cette étude cherche à comprendre les mécanismes d’harmonisation physiologique de la réponse de stress chez les dyades. Les données qui ont servi à ce projet de mémoire proviennent d’une étude longitudinale de 1995 qui portait sur des mères adolescentes. Bien que les résultats des analyses corrélationnelles ne permettent pas d’identifier des différences significatives entre les différents types d’attachement, les résultats des analyses de variances mixtes à mesures répétées suggèrent que chez les dyades mères-enfants, l’harmonisation du cortisol salivaire se développerait de façon significative entre le 9e et 15e mois, et ce, indépendamment du type d’attachement de l’enfant. Ces résultats suggèrent que, dans les contextes de réactivité au stress, l’harmonisation de l’axe HPS des dyades mères-enfants serait un mécanisme physiologique normatif qui se développe à travers le temps. / Attachment security plays a significant role in stress management. Indeed, it allows the child to regulate his behavior, emotions and physiology. The present dissertation, which is part of a research data valuation approach, aims to investigate, from a longitudinal point of view, the link between attachment security and the physiological synchronization of stress mechanisms in mother-child dyads. By studying the reactivity of the hypothalamo-pituitary-adrenal (HPA) axis that regulates cortisol, often referred to as the stress hormone, this study seeks to understand the mechanisms of physiological attunement of the stress response in dyads. The data for this dissertation project came from a 1995 longitudinal study of adolescent mothers. Although the results of the correlational analyses did not identify significant differences between the different types of attachment, the results of the repeated measures analysis of variance using mixed models, suggest that within mother-child dyads, the attunement of salivary cortisol would develop significantly between the 9th and 15th month of age, independently of the child's type of attachment. These results suggest that, in contexts of stress reactivity, the attunement of the HPS axis of mother-infant dyads would be a normative physiological mechanism that develops over time.

Diffuse large B cell lymphoma (DLBCL) is the most common non-Hodgkin’s lymphoma. Patients who fail conventional therapy (~50%) have a poor prognosis and few treatment options. It is essential to understand the underlying biological processes, the progression of the disease, and utilize this information to develop new therapeutics. DLBCL patients with high C-MYC expression have a poor prognosis and new therapeutics for these patients are needed. This thesis describes work testing the hypothesis that JQ1, which can indirectly inhibit C-MYC in some tumors, can be used as an effective treatment for DLBCL. Some tumors have an unknown mechanism causing high C-MYC expression, leading me to investigate the underlying mechanisms. YY1 is a transcriptional regulator of c- Myc and has been implicated in DLBCL and as a potential regulator of the germinal center (GC) reaction. DLBCL arises from GC cells or post-GC cells. I tested the hypothesis that YY1 regulates the GC reaction. SMURF2 is an E3-ubiquitin ligase for YY1 and a tumor suppressor for DLBCL. I was interested in examining the mechanism underlying the suppression of DLBCL by SMURF2 leading to the hypothesis that SMURF2 regulates the GC. This thesis shows JQ1 leads to cell death and cellular senescence in human DLBCL cells. I conclude that BRD4 inhibition by JQ1 or derivatives could provide a new therapeutic avenue for DLBCL patients. I also show loss of YY1 perturbs the GC by decreasing the dark zone and increasing apoptosis. Finally I show modulation of SMURF2 does not affect the GC, suggesting SMURF2 utilizes a different mechanism to act as a tumor suppressor and may not modulate YY1 in the context of the GC.

Dans cette thèse, nous proposons l'utilisation de représentations de niveau intermédiaire, et en particulier i) d'axes médians, ii) de parties d'objets, et iii) des caractéristiques convolutionnels, pour modéliser des objets.La première partie de la thèse traite de détecter les axes médians dans des images naturelles en couleur. Nous adoptons une approche d'apprentissage, en utilisant la couleur, la texture et les caractéristiques de regroupement spectral pour construire un classificateur qui produit une carte de probabilité dense pour la symétrie. Le Multiple Instance Learning (MIL) nous permet de traiter l'échelle et l'orientation comme des variables latentes pendant l'entraînement, tandis qu'une variante fondée sur les forêts aléatoires offre des gains significatifs en termes de temps de calcul.Dans la deuxième partie de la thèse, nous traitons de la modélisation des objets, utilisant des modèles de parties déformables (DPM). Nous développons une approche « coarse-to-fine » hiérarchique, qui utilise des bornes probabilistes pour diminuer le coût de calcul dans les modèles à grand nombre de composants basés sur HOGs. Ces bornes probabilistes, calculés de manière efficace, nous permettent d'écarter rapidement de grandes parties de l'image, et d'évaluer précisément les filtres convolutionnels seulement à des endroits prometteurs. Notre approche permet d'obtenir une accélération de 4-5 fois sur l'approche naïve, avec une perte minimale en performance.Nous employons aussi des réseaux de neurones convolutionnels (CNN) pour améliorer la détection d'objets. Nous utilisons une architecture CNN communément utilisée pour extraire les réponses de la dernière couche de convolution. Nous intégrons ces réponses dans l'architecture DPM classique, remplaçant les descripteurs HOG fabriqués à la main, et nous observons une augmentation significative de la performance de détection (~14.5% de mAP).Dans la dernière partie de la thèse nous expérimentons avec des réseaux de neurones entièrement convolutionnels pous la segmentation de parties d'objets.Nous réadaptons un CNN utilisé à l'état de l'art pour effectuer une segmentation sémantique fine de parties d'objets et nous utilisons un CRF entièrement connecté comme étape de post-traitement pour obtenir des bords fins.Nous introduirons aussi un à priori sur les formes à l'aide d'une Restricted Boltzmann Machine (RBM), à partir des segmentations de vérité terrain.Enfin, nous concevons une nouvelle architecture entièrement convolutionnel, et l'entraînons sur des données d'image à résonance magnétique du cerveau, afin de segmenter les différentes parties du cerveau humain.Notre approche permet d'atteindre des résultats à l'état de l'art sur les deux types de données. / In this thesis we propose the use of mid-level representations, and in particular i) medial axes, ii) object parts, and iii)convolutional features, for modelling objects.The first part of the thesis deals with detecting medial axes in natural RGB images. We adopt a learning approach, utilizing colour, texture and spectral clustering features, to build a classifier that produces a dense probability map for symmetry. Multiple Instance Learning (MIL) allows us to treat scale and orientation as latent variables during training, while a variation based on random forests offers significant gains in terms of running time.In the second part of the thesis we focus on object part modeling using both hand-crafted and learned feature representations. We develop a coarse-to-fine, hierarchical approach that uses probabilistic bounds for part scores to decrease the computational cost of mixture models with a large number of HOG-based templates. These efficiently computed probabilistic bounds allow us to quickly discard large parts of the image, and evaluate the exact convolution scores only at promising locations. Our approach achieves a $4times-5times$ speedup over the naive approach with minimal loss in performance.We also employ convolutional features to improve object detection. We use a popular CNN architecture to extract responses from an intermediate convolutional layer. We integrate these responses in the classic DPM pipeline, replacing hand-crafted HOG features, and observe a significant boost in detection performance (~14.5% increase in mAP).In the last part of the thesis we experiment with fully convolutional neural networks for the segmentation of object parts.We re-purpose a state-of-the-art CNN to perform fine-grained semantic segmentation of object parts and use a fully-connected CRF as a post-processing step to obtain sharp boundaries.We also inject prior shape information in our model through a Restricted Boltzmann Machine, trained on ground-truth segmentations.Finally, we train a new fully-convolutional architecture from a random initialization, to segment different parts of the human brain in magnetic resonance image data.Our methods achieve state-of-the-art results on both types of data.

Indiana University-Purdue University Indianapolis (IUPUI) / Hypothalamic-pituitary-adrenal (HPA) axis dysfunction is found in heavy alcohol users. Negative urgency is a personality trait reflecting the tendency to act rashly in response to negative emotional states, and is associated with problematic alcohol consumption. The current study examined the relationship between negative urgency and HPA axis functioning following (1) negative mood induction and (2) intravenous alcohol administration among heavy social drinkers (proposed n = 40). I hypothesized the following: (1) Negative mood induction would result in an increase of cortisol release as compared to neutral mood induction; (1a) Negative urgency would be related to increased cortisol release in response to negative mood induction; (1b) Negative urgency would partially mediate the relationship between mood induction and cortisol release; (2) Acute IV alcohol administration would result in increased cortisol levels in the neutral mood condition, but decreased cortisol levels in the negative mood condition; and (2a) Negative urgency would be related to the suppression of cortisol release in the negative mood condition in response to acute IV alcohol administration. Repeated measures analyses of variance, the PROCESS macro, and paired samples t-tests were used to examine study hypotheses. Hypotheses were largely unsupported. Writing mood induction procedures reduced salivary cortisol levels in negative mood (t(35)= 2.49, p= 0.02) and there was a trend decrease in neutral mood (t(35)= 1.87, p= 0.07). Alcohol administration also reduced salivary cortisol levels in both negative mood (t(35)= 3.99, p< 0.01) and neutral mood (t(35)= 2.60, p= 0.01). However, salivary cortisol changes were no different than typical circadian patterns in response to mood induction (t(231)= 0.37, p=0.71) or in response to acute alcohol administration (t(231)= 0.44, p= 0.64). Negative urgency had a trend main effect on salivary cortisol level in response to acute IV alcohol administration, such that those higher in negative urgency were more similar to typical circadian patterns (F(19,28)= 1.59, p=0.13). This could serve as preliminary support for a psychological mechanism for the alcohol sensitivity hypothesis. Overall these findings suggest the current study failed to sufficiently manipulate salivary cortisol levels. Future studies should consider methodological techniques when exploring these relationships, including IV compared to oral alcohol administration, mood compared to stress manipulations, and cortisol compared to other HPA axis biomarkers.

Indiana University-Purdue University Indianapolis (IUPUI) / Prospective epidemiologic evidence indicates that both stressful life events (SLEs) and depression are associated with an increased risk of subclinical atherosclerosis and cardiovascular disease (CVD) events. Even though stressful life events (SLEs) and depression co-occur and may act together to influence cardiovascular disease (CVD) risk, these psychosocial factors have been mainly examined in isolation. For instance, depression may moderate the relationship between SLEs and CVD outcomes. I hypothesized that depressive symptoms would potentiate the deleterious effect of SLEs on subclinical atherosclerosis. This hypothesis is plausible, given that depressed adults exhibit exaggerated and prolonged sympathetic nervous system, hypothalamic-pituitary-adrenal (HPA) axis, and inflammatory responses to stress, which in turn could promote atherosclerosis. As compared to their nondepressed counterparts, depressed individuals may also be more likely to engage in maladaptive methods to cope with SLEs (e.g., increased tobacco use, alcohol use, and consumption of low-nutrient, energy dense foods), which could also promote atherosclerosis. I examined cross-sectional data from 274 to 279 (depending on the outcome measure) older, African American adults (mean age = 66 years, 67% female) with no evidence of clinical CVD or dementia who participated in the St. Louis African American Health-Heart study (2009–2011). Number of SLEs was assessed using the Life Events Calendar, a structured interview. From this interview, a continuous SLEs variable was computed (number of adult SLEs: 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or 11+). Severity of depression symptoms was measured using the 17-item Hamilton Rating Scale for Depression (HAM-D). Two measures of subclinical atherosclerosis were obtained: carotid intima-media thickness (CIMT; assessed by ultrasonography) and coronary artery calcification (CAC; assessed by multi-detector computerized tomography). I conducted linear (CIMT) and logistic (CAC) regression models, first adjusted for demographics (age, sex, education) and then fully-adjusted (demographics; mean arterial pressure; low density lipoprotein cholesterol (LDL-C), high density lipoprotein cholesterol (HDL-C); hemoglobin A1c; BMI; tobacco use; diabetes diagnosis; and use of antihypertensitve, lipid lowering, antidiabetic, and antidepressant medications). No main effects of SLEs or HAM-D were found for CIMT or CAC. There were also no SLEs by HAM-D interactions for CIMT or CAC. Because the current results are largely inconsistent with prior literature and there is a paucity of studies utilizing African American samples, future research is needed to examine the independent and interactive associations of SLEs and depressive symptoms with measures of subclinical atherosclerosis. If the present results are replicated, it may suggest that SLEs, depressive symptoms, and their interactive effect are not cardiotoxic among African American adults.