Search

Supplemental material, sj-docx-2-msj-10.1177_13524585221112605 for Machine learning classification of multiple sclerosis in children using optical coherence tomography by Beyza Ciftci Kavaklioglu, Lauren Erdman, Anna Goldenberg, Can Kavaklioglu, Cara Alexander, Hannah M Oppermann, Amish Patel, Soaad Hossain, Tara Berenbaum, Olivia Yau, Carmen Yea, Mina Ly, Fiona Costello, Jean K Mah, Arun Reginald, Brenda Banwell, Giulia Longoni and E Ann Yeh in Multiple Sclerosis Journal

Supplemental material, sj-docx-1-msj-10.1177_13524585221112605 for Machine learning classification of multiple sclerosis in children using optical coherence tomography by Beyza Ciftci Kavaklioglu, Lauren Erdman, Anna Goldenberg, Can Kavaklioglu, Cara Alexander, Hannah M Oppermann, Amish Patel, Soaad Hossain, Tara Berenbaum, Olivia Yau, Carmen Yea, Mina Ly, Fiona Costello, Jean K Mah, Arun Reginald, Brenda Banwell, Giulia Longoni and E Ann Yeh in Multiple Sclerosis Journal

Supplemental material, sj-docx-1-msj-10.1177_13524585221112605 for Machine learning classification of multiple sclerosis in children using optical coherence tomography by Beyza Ciftci Kavaklioglu, Lauren Erdman, Anna Goldenberg, Can Kavaklioglu, Cara Alexander, Hannah M Oppermann, Amish Patel, Soaad Hossain, Tara Berenbaum, Olivia Yau, Carmen Yea, Mina Ly, Fiona Costello, Jean K Mah, Arun Reginald, Brenda Banwell, Giulia Longoni and E Ann Yeh in Multiple Sclerosis Journal

Supplemental material, sj-docx-2-msj-10.1177_13524585221112605 for Machine learning classification of multiple sclerosis in children using optical coherence tomography by Beyza Ciftci Kavaklioglu, Lauren Erdman, Anna Goldenberg, Can Kavaklioglu, Cara Alexander, Hannah M Oppermann, Amish Patel, Soaad Hossain, Tara Berenbaum, Olivia Yau, Carmen Yea, Mina Ly, Fiona Costello, Jean K Mah, Arun Reginald, Brenda Banwell, Giulia Longoni and E Ann Yeh in Multiple Sclerosis Journal

Supplemental material, sj-docx-3-msj-10.1177_13524585221112605 for Machine learning classification of multiple sclerosis in children using optical coherence tomography by Beyza Ciftci Kavaklioglu, Lauren Erdman, Anna Goldenberg, Can Kavaklioglu, Cara Alexander, Hannah M Oppermann, Amish Patel, Soaad Hossain, Tara Berenbaum, Olivia Yau, Carmen Yea, Mina Ly, Fiona Costello, Jean K Mah, Arun Reginald, Brenda Banwell, Giulia Longoni and E Ann Yeh in Multiple Sclerosis Journal

Using a novel trait-based measure, we examined genetic variants associated with obsessive-compulsive (OC) traits and tested whether OC traits and obsessive-compulsive disorder (OCD) shared genetic risk. We conducted a genome-wide association analysis (GWAS) of OC traits using the Toronto Obsessive-Compulsive Scale (TOCS) in 5018 unrelated Caucasian children and adolescents from the community (Spit for Science sample). We tested the hypothesis that genetic variants associated with OC traits from the community would be associated with clinical OCD using a meta-analysis of all currently available OCD cases. Shared genetic risk was examined between OC traits and OCD in the respective samples using polygenic risk score and genetic correlation analyses. A locus tagged by rs7856850 in an intron of PTPRD (protein tyrosine phosphatase δ) was significantly associated with OC traits at the genome-wide significance level (p = 2.48 × 10-8). rs7856850 was also associated with OCD in a meta-analysis of OCD case/control genome-wide datasets (p = 0.0069). The direction of effect was the same as in the community sample. Polygenic risk scores from OC traits were significantly associated with OCD in case/control datasets and vice versa (p's

Additional file 3: Figure S2. Kaplan-Meier survival curve and risk table of the study population before restricting data to 21 days.

Additional file 6: Figure S4. Scaled Schoenfeld residuals plot of reduced consciousness against the transformed time. Solid black line denotes the smoothing spline fit to residuals of the coefficient for variable reduced consciousness (beta(t)), with the dashed lines indicating a ±2 standard error band. The solid black line systematic departures from the horizontal green line, suggesting non-proportional hazards (i.e., time-dependent effect) of reduced consciousness. Red line is the estimated time-dependent coefficient of reduced consciousness (β(t)= 1.37+ 0.14*t) for the Reduced Mortality Model.

Additional file 3: Figure S2. Kaplan-Meier survival curve and risk table of the study population before restricting data to 21 days.

Using a novel trait-based measure, we examined genetic variants associated with obsessive-compulsive (OC) traits and tested whether OC traits and obsessive-compulsive disorder (OCD) shared genetic risk. We conducted a genome-wide association analysis (GWAS) of OC traits using the Toronto Obsessive-Compulsive Scale (TOCS) in 5018 unrelated Caucasian children and adolescents from the community (Spit for Science sample). We tested the hypothesis that genetic variants associated with OC traits from the community would be associated with clinical OCD using a meta-analysis of all currently available OCD cases. Shared genetic risk was examined between OC traits and OCD in the respective samples using polygenic risk score and genetic correlation analyses. A locus tagged by rs7856850 in an intron of PTPRD (protein tyrosine phosphatase δ) was significantly associated with OC traits at the genome-wide significance level (p = 2.48 × 10−8). rs7856850 was also associated with OCD in a meta-analysis of OCD case/control genome-wide datasets (p = 0.0069). The direction of effect was the same as in the community sample. Polygenic risk scores from OC traits were significantly associated with OCD in case/control datasets and vice versa (p’s rg = 0.71, p = 0.062). We report the first validated genome-wide significant variant for OC traits in PTPRD, downstream of the most significant locus in a previous OCD GWAS. OC traits measured in the community sample shared genetic risk with OCD case/control status. Our results demonstrate the feasibility and power of using trait-based approaches in community samples for genetic discovery.

Additional file 1: Table S1. Definitions of daily clinical signs.

Additional file 7: Figure S5. Performance of model-based scores evaluated on selected landmarking days over time. Time-dependent AUC of using risk scores (calculated from Predictive Model 2: Daily Score) assessed on a specific day (admission, day 2, 5, 7, 10) to predict survival outcome for the subsequent days (including the score day) up to 15 days since admission. AUC=0.5 implies performance is no better than random chance.

Additional file 5: Figure S3. Dynamics in the individual clinical signs and survival outcome (conditional density plots).

Additional file 2: Figure S1. Illustration C-index computation for time-static and time-updated predictions based on survival status of hypothetical subjects. (a) Concordance of Time-static Prediction. Illustration of concordance computation in the scenario where risk assessment is made once at a single time point (e.g., admission) for ultimate survival outcome prediction. (b) Concordance of Time-updated Prediction. Illustration of concordance computation in the scenario where risk assessment is repeated every day and the updated risk score is used for survival prediction. Survival information pertaining to the five hypothetical subjects is colored in blue; filled triangles denote occurrence of death (event), empty triangles denote occurrence of discharge (censoring), empty circles denote subjects remaining at risk at a given time point. Solid black arrows indicate valid pairs of risk score comparisons contributing to concordance computation.

Additional file 9: Figure S6. Performance of day-specific risk scores of different predictive models in predicting survival outcome in the subsequent 2 days. AUC of risk scores at different score days (admission, day 1, day 2, …, day 15) calculated from the three predictive models in discriminating deaths for the subsequent 2 days (including the score day). AUC=0.5 implies performance is no better than random chance.

Additional file 4: Table S2. Missing data on daily clinical warning signs.

Additional file 7: Figure S5. Performance of model-based scores evaluated on selected landmarking days over time. Time-dependent AUC of using risk scores (calculated from Predictive Model 2: Daily Score) assessed on a specific day (admission, day 2, 5, 7, 10) to predict survival outcome for the subsequent days (including the score day) up to 15 days since admission. AUC=0.5 implies performance is no better than random chance.

Additional file 4: Table S2. Missing data on daily clinical warning signs.

Additional file 5: Figure S3. Dynamics in the individual clinical signs and survival outcome (conditional density plots).

Additional file 8: Table S3. Proportion of children with changes in number of CWS between two consecutive hospitalization days.

Additional file 10: Table S4. Patient characteristics at admission, by country.

Additional file 6: Figure S4. Scaled Schoenfeld residuals plot of reduced consciousness against the transformed time. Solid black line denotes the smoothing spline fit to residuals of the coefficient for variable reduced consciousness (beta(t)), with the dashed lines indicating a ±2 standard error band. The solid black line systematic departures from the horizontal green line, suggesting non-proportional hazards (i.e., time-dependent effect) of reduced consciousness. Red line is the estimated time-dependent coefficient of reduced consciousness (β(t)= 1.37+ 0.14*t) for the Reduced Mortality Model.

Additional file 9: Figure S6. Performance of day-specific risk scores of different predictive models in predicting survival outcome in the subsequent 2 days. AUC of risk scores at different score days (admission, day 1, day 2, …, day 15) calculated from the three predictive models in discriminating deaths for the subsequent 2 days (including the score day). AUC=0.5 implies performance is no better than random chance.

Additional file 8: Table S3. Proportion of children with changes in number of CWS between two consecutive hospitalization days.

Anorexia nervosa (AN) and obsessive-compulsive disorder (OCD) are often comorbid and likely to share genetic risk factors. Hence, we examine their shared genetic background using a cross-disorder GWAS meta-analysis of 3495 AN cases, 2688 OCD cases, and 18,013 controls. We confirmed a high genetic correlation between AN and OCD (rg = 0.49 ± 0.13, p = 9.07 × 10-7) and a sizable SNP heritability (SNP h2 = 0.21 ± 0.02) for the cross-disorder phenotype. Although no individual loci reached genome-wide significance, the cross-disorder phenotype showed strong positive genetic correlations with other psychiatric phenotypes (e.g., rg = 0.36 with bipolar disorder and 0.34 with neuroticism) and negative genetic correlations with metabolic phenotypes (e.g., rg = -0.25 with body mass index and -0.20 with triglycerides). Follow-up analyses revealed that although AN and OCD overlap heavily in their shared risk with other psychiatric phenotypes, the relationship with metabolic and anthropometric traits is markedly stronger for AN than for OCD. We further tested whether shared genetic risk for AN/OCD was associated with particular tissue or cell-type gene expression patterns and found that the basal ganglia and medium spiny neurons were most enriched for AN-OCD risk, consistent with neurobiological findings for both disorders. Our results confirm and extend genetic epidemiological findings of shared risk between AN and OCD and suggest that larger GWASs are warranted.

Similarity Network Fusion (SNF) is a novel methodological tool that integrates multiple different types of data to identify homogeneous subsets of patients in whom disease classification may be otherwise unclear or challenging. In this review article, the authors hope to provide insight into how SNF can be used in clinical decision making where the aim is to have little influence on the data prior to obtaining the results of the analysis.

More than two thirds of mental health problems have their onset during childhood or adolescence. Identifying children at risk for mental illness later in life and predicting the type of illness is not easy. We set out to develop a platform to define subtypes of childhood social-emotional development using longitudinal, multifactorial trait-based measures. Subtypes discovered through this study could ultimately advance psychiatric knowledge of the early behavioural signs of mental illness. To this extent we have examined two types of models: latent class mixture models and GP-based models. Our findings indicate that while GP models come close in accuracy of predicting future trajectories, LCMMs predict the trajectories as well in a fraction of the time. Unfortunately, neither of the models are currently accurate enough to lead to immediate clinical impact. The available data related to the development of childhood mental health is often sparse with only a few time points measured and require novel methods with improved efficiency and accuracy., extended abstract for ML4HC at NIPS 2016, 4 pages

Two obsessive-compulsive disorder (OCD) genome-wide association studies (GWASs) have been published by independent OCD consortia, the International Obsessive-Compulsive Disorder Foundation Genetics Collaborative (IOCDF-GC) and the OCD Collaborative Genetics Association Study (OCGAS), but many of the top-ranked signals were supported in only one study. We therefore conducted a meta-analysis from the two consortia, investigating a total of 2688 individuals of European ancestry with OCD and 7037 genomically matched controls. No single-nucleotide polymorphisms (SNPs) reached genome-wide significance. However, in comparison with the two individual GWASs, the distribution of P-values shifted toward significance. The top haplotypic blocks were tagged with rs4733767 (P=7.1 × 10(-7); odds ratio (OR)=1.21; confidence interval (CI): 1.12-1.31, CASC8/CASC11), rs1030757 (P=1.1 × 10(-6); OR=1.18; CI: 1.10-1.26, GRID2) and rs12504244 (P=1.6 × 10(-6); OR=1.18; CI: 1.11-1.27, KIT). Variants located in or near the genes ASB13, RSPO4, DLGAP1, PTPRD, GRIK2, FAIM2 and CDH20, identified in linkage peaks and the original GWASs, were among the top signals. Polygenic risk scores for each individual study predicted case-control status in the other by explaining 0.9% (P=0.003) and 0.3% (P=0.0009) of the phenotypic variance in OCGAS and the European IOCDF-GC target samples, respectively. The common SNP heritability in the combined OCGAS and IOCDF-GC sample was estimated to be 0.28 (s.e.=0.04). Strikingly, ∼65% of the SNP-based heritability in the OCGAS sample was accounted for by SNPs with minor allele frequencies of ⩾40%. This joint analysis constituting the largest single OCD genome-wide study to date represents a major integrative step in elucidating the genetic causes of OCD.Molecular Psychiatry advance online publication, 1 August 2017; doi:10.1038/mp.2017.154.