PreprintArticleVersion 1Preserved in Portico This version is not peer-reviewed
Construction of an Exposure-Pathway-Phenotype in Children with Depression due to Transfusion-Dependent Thalassemia: Results of (Un)supervised Machine Learning
Version 1
: Received: 15 June 2020 / Approved: 16 June 2020 / Online: 16 June 2020 (08:51:33 CEST)
How to cite:
Al-Hakeim, H.; Najm, A.; Moustafa, S.; Maes, M. Construction of an Exposure-Pathway-Phenotype in Children with Depression due to Transfusion-Dependent Thalassemia: Results of (Un)supervised Machine Learning. Preprints2020, 2020060202. https://doi.org/10.20944/preprints202006.0202.v1
Al-Hakeim, H.; Najm, A.; Moustafa, S.; Maes, M. Construction of an Exposure-Pathway-Phenotype in Children with Depression due to Transfusion-Dependent Thalassemia: Results of (Un)supervised Machine Learning. Preprints 2020, 2020060202. https://doi.org/10.20944/preprints202006.0202.v1
Al-Hakeim, H.; Najm, A.; Moustafa, S.; Maes, M. Construction of an Exposure-Pathway-Phenotype in Children with Depression due to Transfusion-Dependent Thalassemia: Results of (Un)supervised Machine Learning. Preprints2020, 2020060202. https://doi.org/10.20944/preprints202006.0202.v1
APA Style
Al-Hakeim, H., Najm, A., Moustafa, S., & Maes, M. (2020). Construction of an Exposure-Pathway-Phenotype in Children with Depression due to Transfusion-Dependent Thalassemia: Results of (Un)supervised Machine Learning. Preprints. https://doi.org/10.20944/preprints202006.0202.v1
Chicago/Turabian Style
Al-Hakeim, H., Shatha Moustafa and Michael Maes. 2020 "Construction of an Exposure-Pathway-Phenotype in Children with Depression due to Transfusion-Dependent Thalassemia: Results of (Un)supervised Machine Learning" Preprints. https://doi.org/10.20944/preprints202006.0202.v1
Abstract
Transfusion dependent thalassemia (TDT) patients are treated with continued blood transfusions and show a higher prevalence of depression. TDT with consequent iron overload and inflammation is associated with increased severity of depressive symptoms in TDT children.Aim of the study: To construct a pathway-phenotype which combines iron overload and neuro-immune biomarkers with depressive symptom subdomains in TDT children.Methods: We measured iron status parameters (iron, ferritin, transferrin saturation percentage) and inflammatory (interleukin-1β and tumour necrosis factor-α) biomarkers in TDT (n=111) and healthy (n=53) children and analyzed the results using machine learning.Results: Cluster analysis separated TDT children with depression from those without depression and revealed two depressive subgroups one with low self-esteem and another with increased social-irritability scores. Exploratory factor analysis validated four depressive symptom dimensions as reliable constructs, namely key depressive, physiosomatic, lowered self-esteem and social-irritability dimensions. Partial Least Squares showed that 73.0% of the variance in a latent vector extracted from those four clinical subdomains, immune-inflammatory and iron overload biomarkers was explained by exposure variables including the number of blood transfusions and hospitalizations and use of deferoxamine. The exposure data, iron and immune biomarkers, and symptom subdomains are reflective manifestations of a single latent trait, which shows internal consistency reliability and predictive relevance.Conclusions: The nomological network combining exposure, pathways and behavioral phenome manifestations provides an index of overall severity and disease risk and, therefore, constitutes a new drug target, indicating that iron overload and immune activation should be targeted to treat depression due to TDT.
Medicine and Pharmacology, Psychiatry and Mental Health
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
