Insulin secretion varies widely in preclinical type 1 diabetes. To understand the pathogenesis of this metabolic heterogeneity, we asked whether genetic predisposition to type 2 diabetes, quantified by a type 2 diabetes genetic risk score (T2D-GRS), modulates beta cell function and disease progression in individuals at risk for type 1 diabetes. We analyzed 4,324 islet-autoantibody-positive TrialNet Pathway to Prevention participants with genome-wide genotyping and oral-glucose-tolerance testing. Both T2D-GRS and the type 1 diabetes genetic risk score-2 (T1D-GRS2) differed significantly across five previously described groups defined by area-under-the-curve (AUC) C-peptide (a measure of insulin secretion). The highest C-peptide group, compared with the lowest, had significantly higher T2D-GRS, lower T1D-GRS2, higher BMI z-score, greater insulin resistance, older age, and lower prevalence of males, multiple islet autoantibody positivity, and IA-2- or insulin-autoantibody positivity. Progression to clinical (Stage 3) type 1 diabetes was significantly associated with T1D-GRS2 across all groups and with T2D-GRS in all but the lowest C-peptide group. In conclusion, type 2 diabetes genetic burden shapes metabolic heterogeneity and accelerates progression in preclinical type 1 diabetes. These results support evaluating type 2 diabetes-related mechanisms as targets to improve the prediction and prevention of type 1 diabetes.

Diabetes increases risk of Coronary Artery Disease (CAD) 2-4x and 68% of all mortality in diabetic patients is attributed to heart disease. Genome-wide association studies (GWAS) have identified >1500 loci independently associated with CAD, T2D, or T1D. Importantly, concordant genomic regions associated across these diseases and shared clinical features suggest convergent genetic architectures across these conditions. Further investigation of the genetic drivers of CAD, T2D, and T1D will unravel the genetic architecture of cardiometabolic disease risk and inform novel therapeutic intervention. Genetic colocalization analysis with approximate bayes factorization was performed on three largest available CAD, T2D, and T1D GWAS summary statistics. Significant loci were mapped to genes ± 10 kb of gene TSS and overlapped with GTEx coronary artery expression quantitative trait loci (eQTL). Multi-marker Analysis of GenoMic Association (MAGMA) was performed for each study. Significant genes were further queried by plaque features (lesion stage, calcification) in a multi-modal single cell atlas of atherosclerosis (MetaPlaq) and plaque expression was associated with molecular features and clinical outcomes in the Athero-Express biobank (n=2,595). We identified significant colocalization for 161 SNPs (158 genes) and 47 SNPs (45 genes) across CAD/ T2D and CAD/ T1D respectively. Of note, we identified 21 genes colocalized across all three traits. Among these, VEGFA colocalized in CAD/ T2D, and was upregulated in advanced atherosclerotic plaques in vascular smooth muscle cells (VSMC) and was associated with upregulation of plaque inflammatory markers (IL6, IL8, slCAM1), increased body fat percentage, and was nominally associated with cardiovascular death in Athero-Express. An INPP5B-associated SNP was colocalized in CAD/ T1D and was identified as a significant coronary artery eQTL in GTEx. INPP5B was also upregulated in monocytes and plaque expression level was correlated with IL6 expression in advanced plaques. INPP5B plaque expression was inversely correlated with incidence of major clinical episodes and cardiovascular death. Together, these analyses identify novel shared features among CAD, T2D, and T1D. Investigation of candidate genes reveal associations with differential expression in relevant cell types, patient features, and clinical outcomes. Ongoing efforts involving functional validation will identify emerging targets with therapeutic benefits across disease.

CONTEXT:Maternal (vs paternal) type 1 diabetes is associated with a relative reduction in type 1 diabetes risk in offspring during early life. OBJECTIVE:To determine whether this effect extends into later life. To clarify the importance of intrauterine exposure to maternal type 1 diabetes, and baseline genetic susceptibility in this context. METHODS:We compared the proportion of individuals with type 1 diabetes diagnosed aged 0 to 88 years of age with affected mothers and fathers across 5 observational studies (n = 11 475), and used random-effects meta-analyses to generate overall effect estimates. We examined this by age at diagnosis, and timing of parental diagnosis relative to offspring birth. We compared the type 1 diabetes genetic risk score (T1D-GRS2) of individuals with affected mothers and fathers. RESULTS:Almost half as many individuals with type 1 diabetes had an affected mother vs father (odds ratio [OR], 0.55; 95% CI, 0.48-0.64; P < .0001). A lower proportion of individuals with affected mothers than fathers was apparent even among individuals diagnosed as adults (>18 years) (OR, 0.63; 95% CI, 0.43-0.91; P = .01). The lower proportion of individuals with maternal vs paternal type 1 diabetes was only observed if maternal diagnosis preceded offspring birth (OR, 0.51; 95% CI, 0.37-0.70; P < .001 vs OR 0.97; 95% CI, 0.69-1.38; P = .87 after birth). T1D-GRS2 was similar between individuals with affected mothers and fathers (P = .25). CONCLUSION:Our analyses suggest intrauterine exposure to maternal type 1 diabetes is associated with long-lasting relative protection against offspring type 1 diabetes, which is independent of genetic susceptibility as measured by T1D-GRS2.

Objective The TrialNet Oral Insulin Prevention Trial (TN07) tested oral insulin to prevent Stage 3 type 1 diabetes in 560 Stage 1 relatives of individuals with type 1 diabetes (T1D). Of the three pre-defined risk strata, participants in Secondary Stratum 1 (SS1), characterized by low first-phase insulin release (n=55), responded significantly better to oral insulin. We aimed to identify genetic factors associated with treatment response. Research Design and Methods The TEDDY-T1DExomeChip was used to genotype 552 participants with available DNA. Cox models examined associations between response to oral insulin and HLA haplotypes, 33 pre-selected T1D-associated SNPs, the T1D genetic risk score-2 (T1D-GRS2), and type 2 diabetes (T2D)-associated polygenic scores. For primary analyses, p-values were Benjamini-Hochberg (BH)-corrected for multiple comparisons; results not passing correction were considered nominal. Results GLIS3 rs7020673 was significantly associated with response to oral insulin in SS1 (BH-corrected p-value=0.031 without and p=0.022 with covariate adjustment). Additional nominal associations included better response with HLA-DRB1*04:01-DQA1*03:01-DQB1*03:02 (HR=0.22 vs HR=1.09; unadjusted/adjusted p=0.031/0.045) in SS1, and worse responses with TNFAIP3 and CTLA4 in at least one stratum. In exploratory analyses, participants with T1D-GRS2 >12.5 responded better to oral insulin (HR=0.68) than those with T1D-GRS2 ≤12.5 (HR=2.10; unadjusted/adjusted p=0.003/0.006) in the overall cohort, and lower proinsulin- and obesity-partitioned T2D polygenic scores were associated with greater treatment benefit in SS1 and in another secondary stratum, respectively. Conclusions Genetic differences distinguish responders from non-responders to oral insulin for T1D prevention. Genetics may enable precision medicine by identifying individuals likely to benefit from T1D-modifying therapies.

Genome-wide studies have identified significant allelic associations between genetic variants in or near the IKZF1 gene and multiple autoimmune disorders. IKZF1, encoding the transcription factor IKAROS, produces at least 10 distinct transcripts. To explore the impact of alternative splicing of IKZF1 on the function of mature T cells and the risk of autoimmunity, we generated a panel of human T-cell clones with truncating mutations in IKZF1 exons 4, 6, or both. Differences in gene expression, chromatin accessibility, and protein abundance among clones were assessed by RNA-seq, ATAC-seq, and immunoblotting. Clones with single targeting events clustered separately from double-targeted clones on multiple parameters, but overall, clone responses were highly heterogeneous. Perturbation of IKZF1 splicing resulted in significant differences in expression and chromatin accessibility of other autoimmunity-associated genes and elicited compensatory expression changes in other IKAROS family members. Our results suggest that even modest alterations of IKZF1 splicing can have significant effects on gene expression and function in mature T cells, potentially contributing to autoimmunity in susceptible individuals.

Introduction and Objective: In population screening for islet autoantibodies (IAbs) and type 1 diabetes (T1D) risk, most individuals tested ‘positive’ have only one detectable IAb. Single IAb by standard radio binding assay (RBA) has limited prediction for progression to stage 3 T1D. Combining electrochemiluminescence (ECL) antibody assays with T1D genetic risk score (GRS) may improve prediction. This study examined ECL IAbs, T1D GRS2, and their combined predictive value in TrialNet participants with a single IAb. Methods: We analyzed first available IAb-positive samples from 853 TrialNet screening study participants who tested positive by RBA for a single IAb confirmed across ≥2 consistent visits (575 with GADA, 236 with IAA, 42 with IA-2A). The samples were blindly tested with ECL IAb assays. All participants were previously genotyped using the TEDDY array (GWAS with exome and custom content) and had T1D GRS2 scores determined. Results: Of those with single IAb by RBA, 323/853 (37.9%) were confirmed with ECL assays, showing a significantly higher 5-year cumulative incidence of stage 3 T1D (14.7% vs. 2.2%, p <0.001) and higher T1D GRS2 scores (12.9 ± 2.4 vs. 11.8 ± 2.3, p <0.001) than those not confirmed by ECL assays. ECL-GADA was significantly correlated with T1D GRS2 for both positivity (r=0.26) and levels (r=0.18, both p <0.0001). Both ECL IAb positivity (HR 8.0, 95% CI 3.7-17.6) and T1D GRS2 (HR 1.3, 95% CI 1.1-1.6) independently predicted progression to stage 3 T1D after adjusting for age, sex and BMI. Using ROC AUC as a metric, combining ECL IAb and T1D GRS2 improved 5-year prediction of stage 3 T1D to AUC 0.816 compared to ECL IAb (AUC 0.806, n.s.) or GRS2 (AUC 0.743, p <0.01) alone. Conclusion: In summary, both ECL-detected IAbs and T1D GRS independently enhance the disease prediction among single IAb positive individuals. Their combination further improves prediction, supporting an integrated approach for improved risk stratification in early-stage T1D screening strategies. Disclosure X. Jia: None. C. Zhang: None. A. Steck: Consultant; Current; Sanofi. R.A. Oram: Other - Research support, The University of Exeter has a licensing and royalty agreement for a 10 SNP T1D GRS with Randox; Current; Randox. Other - Advisory Panel, Consulting, Invited talks; Current; Sanofi. Other - invited talk, internal teaching talk on genetics; Ended; Novo Nordisk. D. Cuthbertson: None. H. Parikh: None. S. Onengut-Gumuscu: None. S. Rich: Advisory Panel; Current; Sanofi. Research Support; Current; Sanofi, Leona M. and Harry B. Helmsley Charitable Trust. Other - Consulting Associate Editor, Diabetes Care; Current; American Diabetes Association. J. Krischer: None. P. Gottlieb: Consultant; Current; Eli Lilly and Company. Board Member; Current; IM Therapeutics. Other - CEO, CMO; Current; IM Therapeutics. Research Support; Current; Immune Tolerance Network, Nova Laboratories, National Institute of Diabetes and Digestive and Kidney Diseases. Advisory Panel; Current; Sanofi. Research Support; Current; Sanofi. Consultant; Current; SAB Biotherapeutics, Inc., Anaptys Bio, Cour, T1D Fund. Consultant; Ended; Imcyse, Viacyte, Abata. M. Redondo: Advisory Panel; Current; Sanofi. Other - Data Safety Monitoring committee; Current; Lilly. L. Yu: None. Funding Colorado Diabetes Research Center (DRC) Pilot and Feasibility (P&F) program

Context: The influence of genetic factors on the transition through preclinical stages of type 1 diabetes (T1D) has not been studied. Objective: Our aim was to evaluate the influence of genetic factors on transition through T1D stages. Methods: In TrialNet participants who have been genotyped with the TEDDY-T1DExomeChip array (Illumina HumanCoreExome Beadarray with custom content), we evaluated the influence of the overall T1D genetic risk score (GRS2), its human leukocyte antigen (HLA) and non-HLA components, HLA-DR3 and HLA-DR4 haplotypes, and 90 single-nucleotide variations previously associated with islet autoimmunity and/or T1D on 3 transitions between diabetes stages: from single confirmed autoantibody positive to stage 1 (N = 4314), from stage 1 to stage 2 (N = 3066), and from stage 2 to stage 3 (clinical) T1D (N = 2045). Results: The T1D GRS2 was associated with all 3 transitions with hazard ratios (HRs) of 1.11 (1.09-1.14) for single-autoantibody positivity to stage 1, HR 1.05 (1.03-1.08) for stage 1 to 2, and HR 1.13 (1.09-1.17) for stage 2 to 3 T1D. The T1D GRS2 HLA and HLA class II components were associated with all 3 transitions. The HLA class I component and the HLA-DR4 haplotype were associated with the transition from single-autoantibody positivity to stage 1 and from stage 2 to stage 3 T1D, while HLA-DR3 was associated only with the latter. Conclusion: Genetics influence transitions through each stage of preclinical T1D, with main contributions from HLA class II. These results increase our understanding of T1D development and support incorporating the T1D GRS2 to enhance the prediction of progression through the preclinical stages of T1D.

Chronic obstructive pulmonary disease (COPD) exhibits marked heterogeneity in lung function decline, mortality, exacerbations, and other disease-related outcomes. Omic risk scores (ORS) estimate the cumulative contribution of omics, such as the transcriptome, proteome, and metabolome, to a particular trait. This study evaluated associations between blood-based ORS and COPD-related traits in both smoking-enriched and general population cohorts. ORS were developed and tested in 3,339 participants of Genetic Epidemiology of COPD (COPDGene) with blood RNA-sequencing, proteomic, and metabolomic data. Single- and multi-omic risk scores were trained on 24 cross-sectional and five longitudinal traits using 80

Polygenic scores strongly predict type 1 diabetes risk, but most scores were developed in European-ancestry populations. In this study, we leveraged recent multi-ancestry genome-wide association studies to create a type 1 diabetes multi-ancestry polygenic score (T1D MAPS). We trained the score in the Mass General Brigham (MGB) Biobank (372 individuals with type 1 diabetes) and tested the score in the All of Us program (86 individuals with type 1 diabetes). We evaluated the area under the receiver operating characteristic curve (AUC), and we compared the AUC to two published single-ancestry scores: T1D GRS2EUR and T1D GRSAFR. We also developed an updated score (T1D MAPS2) that combines T1D GRS2EUR and T1D MAPS. Among individuals with non-European ancestry, the AUC of T1D MAPS was 0.90, significantly higher than T1D GRS2EUR (0.82) and T1D GRSAFR (0.82). Among individuals with European ancestry, the AUC of T1D MAPS was slightly lower than T1D GRS2EUR (0.89 vs. 0.91). However, T1D MAPS2 performed equivalently to T1D GRS2EUR in European ancestry (0.91 vs. 0.91) and performed better in non-European ancestry (0.90 vs. 0.82). Overall, these findings advance the accuracy of type 1 diabetes genetic risk prediction across diverse populations.

Efficient prediction of clinical type 1 diabetes is important for risk stratification and monitoring of autoantibody-positive individuals. In this study, we compared type 1 diabetes predictive models for predictive performance, cost and participant time needed for testing. We developed 1943 predictive models using a Cox model based on a type 1 diabetes genetic risk score (GRS2), autoantibody count and types, BMI, age, self-reported gender and OGTT-derived glucose and C-peptide measures. We trained and validated the models using halves of a dataset comprising autoantibody-positive first-degree relatives of individuals with type 1 diabetes (n=3967, 49

Accurate type 1 diabetes prediction is important to facilitate screening for pre-clinical type 1 diabetes to enable potential early disease-modifying interventions and to reduce the risk of severe presentation with diabetic ketoacidosis. We aimed to assess the generalisability of a prediction model developed in children followed from birth. Additionally, we sought to create an application for easy calculation and visualisation of individualised risk prediction. We developed and internally validated a stratified prediction model combining a genetic risk score, age, islet autoantibodies, and family history using data from children followed since birth by The Environmental Determinants of Diabetes in the Young (TEDDY) study. We tested the validity of the model through external validation in the Type 1 Diabetes TrialNet Pathway to Prevention study, which conducts cross-sectional screening in relatives of people with type 1 diabetes. We recalibrated the model by adjusting for baseline risk and selection criteria in TrialNet using logistic recalibration to improve calibration across all ages. The study included 7798 TEDDY and 4068 TrialNet participants, with 305 (4 https://t1dpredictor.diabetesgenes.org ). A stratified model incorporating the type 1 diabetes genetic risk score, family history, age, and autoantibody status can predict type 1 diabetes risk with improved accuracy, but may need recalibration depending on the screening strategy.

Type 1 diabetes (T1D) results from the autoimmune destruction of the insulin-producing β cells. Genetic factors account for approximately 50% of the risk for T1D but, by the late 1990s, the genetic basis was limited. The Type 1 Diabetes Genetics Consortium (T1DGC) was formed in 2002 to accelerate discovery of genes contributing to T1D risk through a grant from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) to assemble existing data and samples from affected sib-pair families and to establish new collections. In recognition of the 75th anniversary of the NIDDK, this manuscript highlights the contributions made by the T1DGC to understanding the genetic basis of T1D using both family (for linkage) and case-control (for genome-wide association) designs. The T1DGC conducted large-scale genetic research and used fine mapping to define risk regions. The T1DGC data, results, and samples have been made available to the scientific community, leading to the discovery of more than 100 loci associated with T1D risk, many with small effects and relevant to autoimmune pathways. The T1DGC not only expanded the list of genes contributing to disease risk but also identified noncoding genetic variation in disease-relevant cell types that contribute to the etiology of T1D. The success of the T1DGC and the NIDDK investment in the global consortium is highlighted in its continuing effect on mapping genetic variants to their function and identifying pathways that provide new targets for the prediction, prevention, and treatment of T1D.

Introduction and Objective: One-third of chronic pancreatitis (CP) patients who undergo total pancreatectomy with islet autotransplantation (TPIAT) achieve insulin independence. Little is known regarding diabetes (DM) related genetics on TPIAT. Genetic risk scores (GRS) are utilized in type 1 (T1D) and type 2 (T2D) DM to predict C-peptide (Cpep) before diagnosis. Methods: We genotyped 384 patients with CP who underwent TPIAT enrolled in the multicenter “POST” study using the Global Screening Array to assess the influence of T1D and T2D GRS and 14 pre-specified SNPs associated with Cpep preservation on 1 year TPIAT outcomes. Results: Patients were 30 ± 17 years, 62% female, 92% Caucasian, 13% previously had DM and received 4294 ± 3458 IEQ/kg during TPIAT. One year post TPIAT 80% required insulin, fasting Cpep was 0.98 ± 0.79 ng/mL and insulin dose adjusted A1c (IDAA1C) was 8.4 ± 2.8. Having diabetes before TPIAT was associated with higher T1D GRS1 (mean Z-score 0.27 ± 0.8 vs -0.04 ± 1.0 in non-DM, p=0.03) and higher T2D GRS (mean Z-score 0.52 ± 1.2 vs -0.07 ± 1.0, p=0.003). At 1 year after TPIAT higher T2D GRS was associated with a lower fasting Cpep (β=-0.09 per 1 SD, p=0.04) and higher IDAA1C (β=0.45 per 1 SD, p=0.009). Several pre-selected SNPs had a trend of lower mean fasting Cpep 1 year post TPIAT (Table). Conclusion: Genetic risk factors for T1D and T2D may impact the risk for DM in CP and insulin secretion after TPIAT. T.M. Triolo: None. A. Eaton: None. W. Chen: None. S. Onengut-Gumuscu: None. A. Steck: Advisory Panel; Sanofi-Aventis U.S. M. Bellin: Advisory Panel; Vertex Pharmaceuticals Incorporated. Research Support; ViaCyte, Inc. Advisory Panel; Novo Nordisk. Consultant; Soleno. Research Support; Dexcom, Inc. Advisory Panel; bridgebio. NIDDK (K23 DK136931)

Background Diabetic kidney disease (DKD) is a serious diabetes complication caused by both environmental and genetic risk factors. Previous genome-wide association studies (GWAS) have identified several loci associated with kidney function and kidney disease in the general population and, to a lesser extent, in diabetes. Methods To uncover the genetic factors driving diabetes-induced kidney function, we conducted a series of GWAS meta-analyses of eGFR in 17,267 individuals with type 1 diabetes and 35,264 with type 2 diabetes (52,531 total), using multiple well-characterized cohorts of type 1 diabetes DKD and data from the UK Biobank and SUrrogate markers for Micro- and Macrovascular hard end points for Innovative diabetes Tools (SUMMIT) consortium. We further accounted for DKD case/control status, diabetes duration and subtype, body mass index, glycated hemoglobin levels, and the relationship between eGFR and albuminuria. Results GWAS identified 13 loci associated with eGFR (P < 5x10(-8)), with five loci (candidate genes: HIPK3, TRIM5, RORA, ERBB4, and BCL6/LPP) not associated with or were in opposite directions as compared with eGFR in the general population. Four candidate genes (HIPK3, BCL6, LPP, and RORA) demonstrated evidence of differential expression in kidney compartments and cells among subgroups with DKD or diabetes versus controls. Lead single-nucleotide polymorphisms rs8027829 (RORA) and rs76300256 (BCL6/LPP) were methylation quantitative trait loci in whole blood and kidney tissue, respectively, and rs76300256 and its related CpGs all cluster in a kidney enhancer. Conclusions Our integrated approach identified candidate genes with diabetes-specific effects on kidney function.

Background Type 1 diabetes (T1D) is characterized by the autoimmune destruction of the insulin-producing beta cells, and there is no cure yet for the disease. While islet autoantibodies are well-recognized biomarkers that mark the onset of islet autoimmunity (IA) and are predictors of T1D, few additional biomarkers are available to monitor disease progression. Recent studies have reported the involvement of complement system proteins in the initiation and progression of IA in the study of T1D. However, the genetic factors of complement system proteins at the time of triggering of IA is unknown. Results Through complement system protein quantitative trait locus (pQTL) mapping analysis of 170 participants from the Diabetes Autoimmunity Study in the Young (DAISY), we identified 240 statistically significant pQTLs (false discovery rate, FDR < 0.1) from pooled and IA case-stratified analyses. Replication analysis conducted on 385 IA cases from The Environment Determinants of Diabetes in the Young (TEDDY) study confirmed 68 significant (FDR < 0.05) pQTLs in total for C8A, C8B, CFB, C4A, and MBL2. Furthermore, all replicated pQTLs of CFB and C4A were previously reported to be associated with T1D risk. Conclusions We identified and replicated 68 pQTLs for five complement system proteins (C8A, C8B, CFB, C4A, and MBL2) in the young population. Among them, all replicated pQTLs of CFB and C4A are also associated with T1D risk. Our study provides evidence of complement system proteins as potential protein biomarkers underlying the development and progression of T1D.

Biological datasets often consist of thousands or millions of variables, e.g. genetic variants or biomarkers, and when sample sizes are large it is common to find many associated with an outcome of interest, for example, disease risk in a GWAS, at high levels of statistical significance, but with very small effects. The False Discovery Rate (FDR) is used to identify effects of interest based on ranking variables according to their statistical significance. Here, we develop a complementary measure to the FDR, the priorityFDR, that ranks variables by a combination of effect size and significance, allowing further prioritisation among a set of variables that pass a significance or FDR threshold. Applying to the largest GWAS of type 1 diabetes to date (15,573 cases and 158,408 controls), we identified 26 independent genetic associations, including two newly-reported loci, with qualitatively lower priorityFDRs than the remaining 175 signals. We detected putatively causal type 1 diabetes risk genes using Mendelian Randomisation, and found that these were located disproportionately close to low priorityFDR signals (p = 0.005), as were genes in the IL-2 pathway (p = 0.003). Selecting variables on both effect size and significance can lead to improved prioritisation for mechanistic follow-up studies from genetic and other large biological datasets.

Rationale A polygenic score (PGS) summarizes a person’s genetic information in a single number for a trait, and its utility in genomic medicine is well-recognized. Although PGS models have been generated for many traits, they are not broadly available for certain traits due to limited sample sizes in studies of infrequent outcomes. Often, prediction of these not well-studied traits ( e.g. , treatment/drug responses) would be of great clinical utility and have been underutilized due to statistical power limitations. To support more versatile trait prediction, we present a method of developing PGS models that can be used in studies of any size with genome-wide SNP data. Method We first generate thousands of PGSs for each study participant in a given data set using their genome-wide SNP data and public resources. A PGS-wide scan involves evaluating the Area Under the Curve (AUC) of prediction for a binary trait (or the R-squared of association for a quantitative trait) at each PGS. We present two methods for the PGS model development, SECRET-Best, which selects the most predictive PGS from the PGS scan for prediction, and SECRET-WTSUM, which considers a combined score from multiple correlation-pruned PGSs. This algorithm is scalable and implemented in a user-friendly software tool, SECRET (Screen and Evaluate Catalogued Risk scores to Enhance Trait predictions). Results We applied SECRET to a binary outcome (type 1 diabetes [T1D]) and a dataset of 2,100 samples, each with 12 laboratory test-related continuous traits. For nine traits with existing PGSs available in the PGS catalog, eight of the traits were predicted correctly, with the same trait-related PGS identified as the top predictor. We showed that the existing PGS methods had rather limited power to predict trait values in a validation set when only 1,500 samples were used to develop a PGS model, while the SECRET methods were able to maintain the prediction power for under-powered GWAS studies even when the sample size of the study was in hundreds. Conclusion The SECRET methods and tool provide a valuable resource for studies with genomic data but of limited sample sizes. This approach enables systematic development and evaluation of PGS models. ### Competing Interest Statement The authors have declared no competing interest.

Background:Accurate type 1 diabetes prediction is important to facilitate screening for pre-clinical type 1 diabetes to enable potential early disease-modifying interventions and to reduce the risk of severe presentation with diabetic ketoacidosis. We aimed to assess the generalisability of a prediction model developed in children followed from birth. Additionally, we sought to create an application for easy calculation and visualization of individualized risk prediction. Methods:We developed and refined a stratified prediction model combining a genetic risk score, age, islet autoantibodies, and family history using data from children followed since birth by The Environmental Determinants of Diabetes in the Young (TEDDY) study. We tested the validity of the model through external validation in the Type 1 Diabetes TrialNet Pathway to Prevention study, which conducts cross-sectional screening in relatives of people with type 1 diabetes. We recalibrated the model by adjusting for baseline risk and selection criteria in TrialNet using logistic recalibration to improve calibration across all ages. Results:The study included 7,798 TEDDY and 4,068 TrialNet participants, with 305 (4%) and 1,373 (34%) developing type 1 diabetes, respectively. The combined model showed similar discriminative ability in autoantibody-positive individuals across TEDDY and TrialNet (p=0.14), but inferior calibration in TrialNet (Brier score 0.40 [0.38,0.43]). Adjustment for baseline risk and selection criteria in TrialNet using logistic recalibration improved calibration across all ages (Brier score 0.16 [0.14,0.17]; p<0.001). A web calculator was developed to visualise individual risk estimates (https://t1dpredictor.diabetesgenes.org). Conclusions:A stratified model of type 1 diabetes genetic risk score, family history, age, and autoantibody status accurately predicts type 1 diabetes risk, but may need recalibration according to screening stategy.