BackgroundThe ability to predict the disease course of individuals with major depressive disorder (MDD) is essential for optimal treatment planning. Here, we used a data-driven machine learning... Show moreBackgroundThe ability to predict the disease course of individuals with major depressive disorder (MDD) is essential for optimal treatment planning. Here, we used a data-driven machine learning approach to assess the predictive value of different sets of biological data (whole-blood proteomics, lipid metabolomics, transcriptomics, genetics), both separately and added to clinical baseline variables, for the longitudinal prediction of 2-year remission status in MDD at the individual-subject level.MethodsPrediction models were trained and cross-validated in a sample of 643 patients with current MDD (2-year remission n = 325) and subsequently tested for performance in 161 individuals with MDD (2-year remission n = 82).ResultsProteomics data showed the best unimodal data predictions (area under the receiver operating characteristic curve = 0.68). Adding proteomic to clinical data at baseline significantly improved 2-year MDD remission predictions (area under the receiver operating characteristic curve = 0.63 vs. 0.78, p = .013), while the addition of other omics data to clinical data did not yield significantly improved model performance. Feature importance and enrichment analysis revealed that proteomic analytes were involved in inflammatory response and lipid metabolism, with fibrinogen levels showing the highest variable importance, followed by symptom severity. Machine learning models outperformed psychiatrists’ ability to predict 2-year remission status (balanced accuracy = 71% vs. 55%).ConclusionsThis study showed the added predictive value of combining proteomic data, but not other omics data, with clinical data for the prediction of 2-year remission status in MDD. Our results reveal a novel multimodal signature of 2-year MDD remission status that shows clinical potential for individual MDD disease course predictions from baseline measurements. Show less
Kluiver, H. de; Jansen, R.; Penninx, B.W.J.H.; Giltay, E.J.; Schoevers, R.A.; Milaneschi, Y. 2023
Depression shows a metabolomic signature overlapping with that of cardiometabolic conditions. Whether this signature is linked to specific depression profiles remains undetermined. Previous... Show moreDepression shows a metabolomic signature overlapping with that of cardiometabolic conditions. Whether this signature is linked to specific depression profiles remains undetermined. Previous research suggested that metabolic alterations cluster more consistently with depressive symptoms of the atypical spectrum related to energy alterations, such as hyperphagia, weight gain, hypersomnia, fatigue and leaden paralysis. We characterized the metabolomic signature of an “atypical/energy-related” symptom (AES) profile and evaluated its specificity and consistency. Fifty-one metabolites measured using the Nightingale platform in 2876 participants from the Netherlands Study of Depression and Anxiety were analyzed. An ‘AES profile’ score was based on five items of the Inventory of Depressive Symptomatology (IDS) questionnaire. The AES profile was significantly associated with 31 metabolites including higher glycoprotein acetyls (β = 0.13, p = 1.35*10-12), isoleucine (β = 0.13, p = 1.45*10-10), very-low-density lipoproteins cholesterol (β = 0.11, p = 6.19*10-9) and saturated fatty acid levels (β = 0.09, p = 3.68*10-10), and lower high-density lipoproteins cholesterol (β = −0.07, p = 1.14*10-4). The metabolites were not significantly associated with a summary score of all other IDS items not included in the AES profile. Twenty-five AES-metabolites associations were internally replicated using data from the same subjects (N = 2015) collected at 6-year follow-up. We identified a specific metabolomic signature—commonly linked to cardiometabolic disorders—associated with a depression profile characterized by atypical, energy-related symptoms. The specific clustering of a metabolomic signature with a clinical profile identifies a more homogenous subgroup of depressed patients at higher cardiometabolic risk, and may represent a valuable target for interventions aiming at reducing depression’s detrimental impact on health. Show less
For a long time it has been thought that habitation and landscape organisation only changed significantly from the Roman Period onwards. However, many developments were already started long before... Show moreFor a long time it has been thought that habitation and landscape organisation only changed significantly from the Roman Period onwards. However, many developments were already started long before Julius Caesar's Roman armies arrived in the southern Netherlands. The Iron Age landscapes were ordered and structured, contrasting with the still open Bronze Age landscapes. Iron Age people inhabited the same places for generations. At the same time they structured their immediate environment and surroundings resulting in a sustainable organisation and arrangement of the landscape.Recent excavations and (micro-)regional archaeological studies into habitation and landscape organisation, among others in the north-eastern region of the province Noord-Brabant, show that relicts from the past strongly dictated the organisation and structuring of later landscapes. The past in the past formed a guideline (dutch: leidraad) for later (Iron Age) inhabitants.The past can also be a guideline for the design, protection and preservation of contemporary landscapes. This aligns with a trend in which archaeologists are explicitly seeking the connection with present society. Therefore this book ends with a plea for a transition of the Dutch archaeological system in which living heritage can also be a guideline for the present. Show less
Windhorst, R.; Alpaslan, M.; Andrews, S.; Ashcraft, T.; Broadhurst, T.; Coe, D.; ... ; Zitrin, A. 2019
The river area Maaskant and adjacent sand area of Oss, located ‘between’ the current course of the river Meuse and the city Oss, are among the most intensively researched regions in the Netherlands... Show moreThe river area Maaskant and adjacent sand area of Oss, located ‘between’ the current course of the river Meuse and the city Oss, are among the most intensively researched regions in the Netherlands. Extensive archaeological and palynological research provides ample opportunities for an interregional research of the occupation and vegetation history of both areas. This article describes the intertwinement between the Holocene river area and the adjacent Pleistocene sandy soils, to eventually get a first insight of the relation(s) between the inhabitants of both regions in late prehistoric and Early Roman period (3000 BC – 250 AD). Show less
Kaal, S.E.J.; Husson, O.; Dartel, F. van; Hermans, K.; Jansen, R.; Manten-Horst, E.; ... ; Graaf, W.T.A. van der 2018