Genomics is failing on diversity

Seeing the random forest through the decision trees. Supporting learning health systems from histopathology with machine learning models: Challenges and opportunities

This paper discusses some overlooked challenges faced when working with machine learning models for histopathology and presents a novel opportunity to support "Learning Health Systems" with them. Initially, the authors elaborate on these challenges after separating them according to their mitigation strategies: those that need innovative approaches, time, or future technological capabilities and those that require a conceptual reappraisal from a critical perspective. Then, a novel opportunity to support "Learning Health Systems" by integrating hidden information extracted by ML models from digitalized histopathology slides with other healthcare big data is presented.

Unfulfilled Needs in the Detection, Diagnosis, Monitoring, Treatment, and Understanding of Glaucoma in Blacks Globally

Glaucoma is an ophthalmic disorder that affects a significant number of Blacks globally. A leading cause of this condition is an age-related enlargement of the lens and increased intraocular pressure. Although Blacks are affected by glaucoma at a higher rate than their Caucasian counterparts, there remains a lack of emphasis placed on the detection, diagnosis, monitoring, and treatment of glaucoma in this population. Education regarding glaucoma in the African and African American populations is essential to reducing rates of glaucoma-related visual impairment and improving treatment success. In this article, we highlight specific issues and limitations to the management of glaucoma, which affects Blacks at a higher rate. In addition, we also review the backgrounds of Blacks globally and examine historical events that have contributed to financial inequality and wealth/health disparities affecting glaucoma management. Lastly, we suggest reparations and solutions that health care professionals can use to improve glaucoma screening and management.

Increasing Diversity, Equity, Inclusion, and Accessibility in Rare Disease Clinical Trials

Diversity, equity, inclusion, and accessibility (DEIA) are foundational principles for clinical trials and medical research. In rare diseases clinical research, where numbers of participants are already challenged by rarity itself, maximizing inclusion is of particular importance to clinical trial success, as well as ensuring the generalizability and relevance of the trial results to the people affected by these diseases. In this article, we review the medical and gray literature and cite case examples to provide insights into how DEIA can be proactively integrated into rare diseases clinical research. Here, we particularly focus on genetic diversity. While the rare diseases DEIA literature is nascent, it is accelerating as many patient advocacy groups, professional societies, training and educational organizations, researcher groups, and funders are setting intentional strategies to attain DEIA goals moving forward, and to establish metrics to ensure continued improvement. Successful examples in underserved and underrepresented populations are available that can serve as case studies upon which rare diseases clinical research programs can be built. Rare diseases have historically been innovation drivers in basic, translational, and clinical research, and ultimately, all populations benefit from data diversity in rare diseases populations that deliver novel insights and approaches to how clinical research can be performed.

Addressing Disparities in Pediatric Congenital Heart Disease: A Call for Equitable Health Care

While significant progress has been made in reducing disparities within the US health care system, notable gaps remain. This article explores existing disparities within pediatric congenital heart disease care. Congenital heart disease, the most common birth defect and a leading cause of infant death, has garnered substantial attention, revealing certain disparities within the US health care system. Factors such as race, ethnicity, insurance coverage, socioeconomic status, and geographic location are all commonalities that significantly affect health disparities in pediatric congenital heart disease. This comprehensive review sheds light on disparities from diverse perspectives in pediatric care, demonstrates the inequities and inequalities leading to these disparities, presents effective solutions, and issues a call to action for providers, institutions, and the health care system. Recognizing and addressing these disparities is imperative for ensuring equitable care and enhancing the long-term well-being of children affected by congenital heart disease. Implementing robust, evidence-based frameworks that promote responsible and safe interventions is fundamental to enduring change.

Genetic research within Indigenous communities: Engagement opportunities and pathways forward

PURPOSE

Against a historical backdrop of researchers who violated trust through lack of benefit sharing, transparency, and engagement, efforts are underway to develop better approaches for genetic and genomic research with Indigenous communities. To increase engagement, there is a need to understand factors that affect researcher and community collaborations. This study aimed to understand the barriers, challenges, and facilitators of Indigenous Peoples in the United States participating in genetic research.

METHODS

We conducted 42 semistructured interviews with Tribal leaders, clinicians, researchers, policy makers, and Tribal research review board members across the United States to explore perceived risks, benefits, barriers, and facilitators of genetic research participation.

RESULTS

Participants, identifying as Indigenous (88%) or non-Indigenous allies (12%), described their concerns, hesitancy, and fears about genetic research, as well as the roles of trust, transparency, and respect for culture in facilitating partnerships. Previous harms-such as sample and data misuse, stigmatization, or misrepresentation by researchers-revealed strategies for building trust to create more equitable and reciprocal research partnerships.

CONCLUSION

Participants in this study offered strategies for increasing genetic research engagement. The pathway forward should foster transparent research policies and practices to facilitate informed research that supports the needs and priorities of participants, communities, and researchers.

A systematic literature review of the 'commercialisation effect' on public attitudes towards biobank and genomic data repositories

Initiatives that collect and share genomic data to advance health research are widespread and accelerating. Commercial interests in these efforts, while vital, may erode public trust and willingness to provide personal genomic data, upon which these initiatives depend. Understanding public attitudes towards providing genomic data for health research in the context of commercial involvement is critical. A PRISMA-guided search of six online academic databases identified 113 quantitative and qualitative studies using primary data pertaining to public attitudes towards commercial actors in the management, collection, access, and use of biobank and genomic data. The presence of commercial interests yields interrelated public concerns around consent, privacy and data security, trust in science and scientists, benefit sharing, and the ownership and control of health data. Carefully considered regulatory and data governance and access policies are therefore required to maintain public trust and support for genomic health initiatives.

Clinical genetics in breast cancer

As genetic testing becomes increasingly more accessible and more applicable with a broader range of clinical implications, it may also become more challenging for breast cancer providers to remain up-to-date. This review outlines some of the current clinical guidelines and recent literature surrounding germline genetic testing, as well as genomic testing, in the screening, prevention, diagnosis, and treatment of breast cancer, while identifying potential areas of further research.

Systematic Review and Phenome-Wide Scans of Genetic Associations with Vascular Cognitive Impairment

Vascular cognitive impairment (VCI) is a heterogenous form of cognitive impairment that results from cerebrovascular disease. It is a result of both genetic and non-genetic factors. Although much research has been conducted on the genetic contributors to other forms of cognitive impairment (e.g. Alzheimer's disease), knowledge is lacking on the genetic factors associated with VCI. A better understanding of the genetics of VCI will be critical for prevention and treatment. To begin to fill this gap, the genetic contributors are reviewed with VCI from the literature. Phenome-wide scans of the identified genes are conducted and genetic variants identified in the review in large-scale resources displaying genetic variant-trait association information. Gene set are also carried out enrichment analysis using the genes identified from the review. Thirty one articles are identified meeting the search criteria and filters, from which 107 unique protein-coding genes are noted related to VCI. The phenome-wide scans and gene set enrichment analysis identify pathways associated with a diverse set of biological systems. This results indicate that genes with evidence of involvement in VCI are involved in a diverse set of biological functions. This information can facilitate downstream research to better dissect possible shared biological mechanisms for future therapies.

Genomic Exploration of Essential Hypertension in African-Brazilian Quilombo Populations: A Comprehensive Approach with Pedigree Analysis and Family-Based Association Studies

Essential Hypertension (EH) is a major global health concern, causing about 9.4 million deaths annually. Its prevalence varies across different regions, affecting 17% of the population in the Americas, 19.2% in the Western Pacific, 23.2% in Europe, 25.1% in Southeast Asia, 26.3% in the Eastern Mediterranean, and 27.2% in Africa. EH is a multifactorial disease influenced by both genetic and environmental factors. While genetic factors contribute 30-60% to blood pressure variation, the genetic complexity of EH remains largely unexplained due to limited knowledge of candidate genes and population-specific differences. Various methods, including candidate gene studies, genome-wide linkage analysis (GWLA), and genome-wide association studies (GWAS), have been employed to identify genetic factors, yet much of the heritability of EH is still unknown. This study aimed to investigate the genetic basis of EH by mapping regions of interest (ROIs) and identifying candidate genes and variants influencing EH in African-derived individuals from partially isolated populations of quilombo remnants in Vale do Ribeira, São Paulo, Brazil. Samples from 431 individuals (167 affected, 261 unaffected, 3 with unknown phenotype) from eight quilombo remnant populations were genotyped using a 650k SNP array. The global ancestry proportions were estimated at 47% African, 36% European, and 16% Native American. Genealogical information from 673 individuals was used to construct six pedigrees comprising 1104 individuals. The mapping strategy consisted of a multi-level computational approach. We constructed pedigrees based on interviews and kinship coefficient, pruned the dataset to obtain three non-overlapping markers subpanels, phased the haplotype and performed local ancestry to account for admixture. We performed GWLA and dense linkage analyses using markers subpanels and performed fine-mapping using family-based association studies (FBAS) based on population and pedigree imputed data, investigating EH-related genes and variants. The linkage analysis identified 22 ROIs with LOD scores 1.45-3.03, containing markers co-segregating with the phenotype. These ROIs encompassed 2363 genes. Fine-mapping identified 60 EH-related candidate genes and 118 suggestive or significant variants (FBAS). Among these, 14 genes, including PHGDH, S100A10, MFN2, and RYR2, were highlighted with strong evidence of association with hypertension. These genes, harboring 29 SNPs, were implicated in regulating blood pressure, sodium and potassium levels, and the aldosterone pathway. This study revealed, through a complementary approach - combining admixture-adjusted genome-wide linkage analysis based on Markov chain Monte Carlo (MCMC) methods, association studies on imputed data, and in silico investigations - genetic regions, variants and candidate genes that shed light on the genetic basis of essential hypertension, with significant potential to explain the genetic etiology in quilombo remnant populations.

Disparities in ABO blood type determination across diverse ancestries: a systematic review and validation in the All of Us Research Program

OBJECTIVES

ABO blood types have widespread clinical use and robust associations with disease. The purpose of this study is to evaluate the portability and suitability of tag single-nucleotide polymorphisms (tSNPs) used to determine ABO alleles and blood types across diverse populations in published literature.

MATERIALS AND METHODS

Bibliographic databases were searched for studies using tSNPs to determine ABO alleles. We calculated linkage between tSNPs and functional variants across inferred continental ancestry groups from 1000 Genomes. We compared r2 across ancestry and assessed real-world consequences by comparing tSNP-derived blood types to serology in a diverse population from the All of Us Research Program.

RESULTS

Linkage between functional variants and O allele tSNPs was significantly lower in African (median r2 = 0.443) compared to East Asian (r2 = 0.946, P = 1.1 × 10-5) and European (r2 = 0.869, P = .023) populations. In All of Us, discordance between tSNP-derived blood types and serology was high across all SNPs in African ancestry individuals and linkage was strongly correlated with discordance across all ancestries (ρ = -0.90, P = 3.08 × 10-23).

DISCUSSION

Many studies determine ABO blood types using tSNPs. However, tSNPs with low linkage disequilibrium promote misinference of ABO blood types, particularly in diverse populations. We observe common use of inappropriate tSNPs to determine ABO blood type, particularly for O alleles and with some tSNPs mistyping up to 58% of individuals.

CONCLUSION

Our results highlight the lack of transferability of tSNPs across ancestries and potential exacerbation of disparities in genomic research for underrepresented populations. This is especially relevant as more diverse cohorts are made publicly available.

TDC-2: Multimodal Foundation for Therapeutic Science

Therapeutics Data Commons (tdcommons.ai) is an open science initiative with unified datasets, AI models, and benchmarks to support research across therapeutic modalities and drug discovery and development stages. The Commons 2.0 (TDC-2) is a comprehensive overhaul of Therapeutic Data Commons to catalyze research in multimodal models for drug discovery by unifying single-cell biology of diseases, biochemistry of molecules, and effects of drugs through multimodal datasets, AI-powered API endpoints, new multimodal tasks and model frameworks, and comprehensive benchmarks. TDC-2 introduces over 1,000 multimodal datasets spanning approximately 85 million cells, pre-calculated embeddings from 5 state-of-the-art single-cell models, and a biomedical knowledge graph. TDC-2 drastically expands the coverage of ML tasks across therapeutic pipelines and 10+ new modalities, spanning but not limited to single-cell gene expression data, clinical trial data, peptide sequence data, peptidomimetics protein-peptide interaction data regarding newly discovered ligands derived from AS-MS spectroscopy, novel 3D structural data for proteins, and cell-type-specific protein-protein interaction networks at single-cell resolution. TDC-2 introduces multimodal data access under an API-first design using the model-view-controller paradigm. TDC-2 introduces 7 novel ML tasks with fine-grained biological contexts: contextualized drug-target identification, single-cell chemical/genetic perturbation response prediction, protein-peptide binding affinity prediction task, and clinical trial outcome prediction task, which introduce antigen-processing-pathway-specific, cell-type-specific, peptide-specific, and patient-specific biological contexts. TDC-2 also releases benchmarks evaluating 15+ state-of-the-art models across 5+ new learning tasks evaluating models on diverse biological contexts and sampling approaches. Among these, TDC-2 provides the first benchmark for context-specific learning. TDC-2, to our knowledge, is also the first to introduce a protein-peptide binding interaction benchmark.

Genomic Insights for Personalized Care: Motivating At-Risk Individuals Toward Evidence-Based Health Practices

Background

Lung cancer and tobacco use pose significant global health challenges, necessitating a comprehensive translational roadmap for improved prevention strategies. Polygenic risk scores (PRSs) are powerful tools for patient risk stratification but have not yet been widely used in primary care for lung cancer, particularly in diverse patient populations.

Methods

We propose the GREAT care paradigm, which employs PRSs to stratify disease risk and personalize interventions. We developed PRSs using large-scale multi-ancestry genome-wide association studies and standardized PRS distributions across all ancestries. We applied our PRSs to 796 individuals from the GISC Trial, 350,154 from UK Biobank (UKBB), and 210,826 from All of Us Research Program (AoU), totaling 561,776 individuals of diverse ancestry.

Results

Significant odds ratios (ORs) for lung cancer and difficulty quitting smoking were observed in both UKBB and AoU. For lung cancer, the ORs for individuals in the highest risk group (top 20% versus bottom 20%) were 1.85 (95% CI: 1.58 - 2.18) in UKBB and 2.39 (95% CI: 1.93 - 2.97) in AoU. For difficulty quitting smoking, the ORs (top 33% versus bottom 33%) were 1.36 (95% CI: 1.32 - 1.41) in UKBB and 1.32 (95% CI: 1.28 - 1.36) in AoU.

Conclusion

Our PRS-based intervention model leverages large-scale genetic data for robust risk assessment across populations. This model will be evaluated in two cluster-randomized clinical trials aimed at motivating health behavior changes in high-risk patients of diverse ancestry. This pioneering approach integrates genomic insights into primary care, promising improved outcomes in cancer prevention and tobacco treatment.

To weight or not to weight? The effect of selection bias in 3 large electronic health record-linked biobanks and recommendations for practice

OBJECTIVES

To develop recommendations regarding the use of weights to reduce selection bias for commonly performed analyses using electronic health record (EHR)-linked biobank data.

MATERIALS AND METHODS

We mapped diagnosis (ICD code) data to standardized phecodes from 3 EHR-linked biobanks with varying recruitment strategies: All of Us (AOU; n = 244 071), Michigan Genomics Initiative (MGI; n = 81 243), and UK Biobank (UKB; n = 401 167). Using 2019 National Health Interview Survey data, we constructed selection weights for AOU and MGI to represent the US adult population more. We used weights previously developed for UKB to represent the UKB-eligible population. We conducted 4 common analyses comparing unweighted and weighted results.

RESULTS

For AOU and MGI, estimated phecode prevalences decreased after weighting (weighted-unweighted median phecode prevalence ratio [MPR]: 0.82 and 0.61), while UKB estimates increased (MPR: 1.06). Weighting minimally impacted latent phenome dimensionality estimation. Comparing weighted versus unweighted phenome-wide association study for colorectal cancer, the strongest associations remained unaltered, with considerable overlap in significant hits. Weighting affected the estimated log-odds ratio for sex and colorectal cancer to align more closely with national registry-based estimates.

DISCUSSION

Weighting had a limited impact on dimensionality estimation and large-scale hypothesis testing but impacted prevalence and association estimation. When interested in estimating effect size, specific signals from untargeted association analyses should be followed up by weighted analysis.

CONCLUSION

EHR-linked biobanks should report recruitment and selection mechanisms and provide selection weights with defined target populations. Researchers should consider their intended estimands, specify source and target populations, and weight EHR-linked biobank analyses accordingly.

Effect of apolipoprotein E ε4 and its modification by sociodemographic characteristics on cognitive measures in South Asians from LASI-DAD

BACKGROUND

We investigated the effects of apolipoprotein E (APOE) ε4 and its interactions with sociodemographic characteristics on cognitive measures in South Asians from the Diagnostic Assessment of Dementia for the Longitudinal Aging Study of India (LASI-DAD).

METHODS

Linear regression was used to assess the association between APOE ε4 and global- and domain-specific cognitive function in 2563 participants (mean age 69.6 ± 7.3 years; 53% female). Effect modification by age, sex, and education were explored using interaction terms and subgroup analyses.

RESULTS

APOE ε4 was inversely associated with most cognitive measures (p < 0.05). This association was stronger with advancing age for the Hindi Mental State Examination (HMSE) score (βε4×age = -0.44, p = 0.03), orientation (βε4×age = -0.07, p = 0.01), and language/fluency (βε4×age = -0.07, p = 0.01), as well as in females for memory (βε4×male = 0.17, p = 0.02) and language/fluency (βε4×male = 0.12, p = 0.03).

DISCUSSION

APOE ε4 is associated with lower cognitive function in South Asians from India, with a more pronounced impact observed in females and older individuals.

HIGHLIGHTS

APOE ε4 carriers had lower global and domain-specific cognitive performance. Females and older individuals may be more susceptible to ε4 effects. For most cognitive measures, there was no interaction between ε4 and education.

The somatic mutation profile of estrogen receptor-positive HER2-negative metastatic breast cancer in Brazilian patients

Background

Breast cancer is the leading cause of cancer death among women worldwide. Studies about the genomic landscape of metastatic breast cancer (MBC) have predominantly originated from developed nations. There are still limited data on the molecular epidemiology of MBC in low- and middle-income countries. This study aims to evaluate the prevalence of mutations in the PI3K-AKT pathway and other actionable drivers in estrogen receptor (ER)+/HER2- MBC among Brazilian patients treated at a large institution representative of the nation's demographic diversity.

Methods

We conducted a retrospective observational study using laboratory data (OC Precision Medicine). Our study included tumor samples from patients with ER+/HER2- MBC who underwent routine tumor testing from 2020 to 2023 and originated from several Brazilian centers within the Oncoclinicas network. Two distinct next-generation sequencing (NGS) assays were used: GS Focus (23 genes, covering PIK3CA, AKT1, ESR1, ERBB2, BRCA1, BRCA2, PALB2, TP53, but not PTEN) or GS 180 (180 genes, including PTEN, tumor mutation burden [TMB] and microsatellite instability [MSI]).

Results

Evaluation of tumor samples from 328 patients was undertaken, mostly (75.6%) with GS Focus. Of these, 69% were primary tumors, while 31% were metastatic lesions. The prevalence of mutations in the PI3K-AKT pathway was 39.3% (95% confidence interval, 33% to 43%), distributed as 37.5% in PIK3CA and 1.8% in AKT1. Stratification by age revealed a higher incidence of mutations in this pathway among patients over 50 (44.5% vs 29.1%, p=0.01). Among the PIK3CA mutations, 78% were canonical (included in the alpelisib companion diagnostic non-NGS test), while the remaining 22% were characterized as non-canonical mutations (identifiable only by NGS test). ESR1 mutations were detected in 6.1%, exhibiting a higher frequency in metastatic samples (15.1% vs 1.3%, p=0.003). Additionally, mutations in BRCA1, BRCA2, or PALB2 were identified in 3.9% of cases, while mutations in ERBB2 were found in 2.1%. No PTEN mutations were detected, nor were TMB high or MSI cases.

Conclusion

We describe the genomic landscape of Brazilian patients with ER+/HER2- MBC, in which the somatic mutation profile is comparable to what is described in the literature globally. These data are important for developing precision medicine strategies in this scenario, as well as for health systems management and research initiatives.

Developing an optimal stratification model for colorectal cancer screening and reducing racial disparities in multi-center population-based studies

BACKGROUND

Early detection of colorectal neoplasms can reduce the colorectal cancer (CRC) burden by timely intervention for high-risk individuals. However, effective risk prediction models are lacking for personalized CRC early screening in East Asian (EAS) population. We aimed to develop, validate, and optimize a comprehensive risk prediction model across all stages of the dynamic adenoma-carcinoma sequence in EAS population.

METHODS

To develop precision risk-stratification and intervention strategies, we developed three trans-ancestry PRSs targeting colorectal neoplasms: (1) using 148 previously identified CRC risk loci (PRS148); (2) SNPs selection from large-scale meta-analysis data by clumping and thresholding (PRS183); (3) PRS-CSx, a Bayesian approach for genome-wide risk prediction (PRSGenomewide). Then, the performance of each PRS was assessed and validated in two independent cross-sectional screening sets, including 4600 patients with advanced colorectal neoplasm, 4495 patients with non-advanced adenoma, and 21,199 normal individuals from the ZJCRC (Zhejiang colorectal cancer set; EAS) and PLCO (the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial; European, EUR) studies. The optimal PRS was further incorporated with lifestyle factors to stratify individual risk and ultimately tested in the PLCO and UK Biobank prospective cohorts, totaling 350,013 participants.

RESULTS

Three trans-ancestry PRSs achieved moderately improved predictive performance in EAS compared to EUR populations. Remarkably, the PRSs effectively facilitated a thorough risk assessment across all stages of the dynamic adenoma-carcinoma sequence. Among these models, PRS183 demonstrated the optimal discriminatory ability in both EAS and EUR validation datasets, particularly for individuals at risk of colorectal neoplasms. Using two large-scale and independent prospective cohorts, we further confirmed a significant dose-response effect of PRS183 on incident colorectal neoplasms. Incorporating PRS183 with lifestyle factors into a comprehensive strategy improves risk stratification and discriminatory accuracy compared to using PRS or lifestyle factors separately. This comprehensive risk-stratified model shows potential in addressing missed diagnoses in screening tests (best NPV = 0.93), while moderately reducing unnecessary screening (best PPV = 0.32).

CONCLUSIONS

Our comprehensive risk-stratified model in population-based CRC screening trials represents a promising advancement in personalized risk assessment, facilitating tailored CRC screening in the EAS population. This approach enhances the transferability of PRSs across ancestries and thereby helps address health disparity.

Effects of Implicit Racial Bias and Standardized Patient Race on Genetic Counseling Students' Patient-Centered Communication

Clinician racial bias has been associated with less patient-centered communication, but little is known about how it affects trainees' communication. We investigated genetic counseling students' communication during sessions with Black or White standardized patients (SPs) and the extent to which communication was associated with SP race or student scores on the Race Implicit Association Test (IAT). Sixty students conducted a baseline SP session and up to two follow-up sessions. Students were randomly assigned to a different White or Black SP and one of three clinical scenarios for each session. Fifty-six students completed the IAT. Session recordings were coded using the Roter Interaction Analysis System. Linear regression models assessed the effects of IAT score and SP race on a variety of patient-centered communication indicators. Random intercept models assessed the within-student effects of SP race on communication outcomes during the baseline session and in follow-up sessions (n = 138). Students were predominantly White (71%). Forty students (71%) had IAT scores indicating some degree of pro-White implicit preference. Baseline sessions with White relative to Black SPs had higher patient-centeredness scores. Within-participant analyses indicate that students used a higher proportion of back-channels (a facilitative behavior that cues interest and encouragement) and conducted longer sessions with White relative to Black SPs. Students' stronger pro-White IAT scores were associated with using fewer other facilitative statements during sessions with White relative to Black SPs. Different patterns of communication associated with SP race and student IAT scores were found for students than those found in prior studies with experienced clinicians.

shaPRS: Leveraging shared genetic effects across traits or ancestries improves accuracy of polygenic scores

We present shaPRS, a method that leverages widespread pleiotropy between traits or shared genetic effects across ancestries, to improve the accuracy of polygenic scores. The method uses genome-wide summary statistics from two diseases or ancestries to improve the genetic effect estimate and standard error at SNPs where there is homogeneity of effect between the two datasets. When there is significant evidence of heterogeneity, the genetic effect from the disease or population closest to the target population is maintained. We show via simulation and a series of real-world examples that shaPRS substantially enhances the accuracy of polygenic risk scores (PRSs) for complex diseases and greatly improves PRS performance across ancestries. shaPRS is a PRS pre-processing method that is agnostic to the actual PRS generation method, and as a result, it can be integrated into existing PRS generation pipelines and continue to be applied as more performant PRS methods are developed over time.

Mapping the Pharmacogenetic Landscape in a Ugandan Population: Implications for Personalized Medicine in an Underrepresented Population

Africans are extremely underrepresented in global genomic research. African populations face high burdens of communicable and non-communicable diseases and experience widespread polypharmacy. As population-specific genetic studies are crucial to understanding unique genetic profiles and optimizing treatments to reduce medication-related complications in this diverse population, the present study aims to characterize the pharmacogenomics profile of a rural Ugandan population. We analyzed low-pass whole genome sequencing data from 1998 Ugandans to investigate 18 clinically actionable pharmacogenes in this population. We utilized PyPGx to identify star alleles (haplotype patterns) and compared allele frequencies across populations using the Pharmacogenomics Knowledgebase PharmGKB. Clinical interpretations of the identified alleles were conducted following established dosing guidelines. Over 99% of participants displayed actionable phenotypes across the 18 pharmacogenes, averaging 3.5 actionable genotypes per individual. Several variant alleles known to affect drug metabolism (i.e., CYP3A5*1, CYP2B6*9, CYP3A5*6, CYP2D6*17, CYP2D6*29, and TMPT*3C)-which are generally more prevalent in African individuals-were notably enriched in the Ugandan cohort, beyond reported frequencies in other African peoples. More than half of the cohort exhibited a predicted impaired drug response associated with CFTR, IFNL3, CYP2B6, and CYP2C19, and approximately 31% predicted altered CYP2D6 metabolism. Potentially impaired CYP2C9, SLCO1B1, TPMT, and DPYD metabolic phenotypes were also enriched in Ugandans compared with other African populations. Ugandans exhibit distinct allele profiles that could impact drug efficacy and safety. Our findings have important implications for pharmacogenomics in Uganda, particularly with respect to the treatment of prevalent communicable and non-communicable diseases, and they emphasize the potential of pharmacogenomics-guided therapies to optimize healthcare outcomes and precision medicine in Uganda.

The VEGF-Hypoxia Signature Is Upregulated in Basal-like Breast Tumors from Women of African Ancestry and Associated with Poor Outcomes in Breast Cancer

PURPOSE

Black women experience the highest breast cancer mortality rate compared with women of other racial/ethnic groups. To gain a deeper understanding of breast cancer heterogeneity across diverse populations, we examined a VEGF-hypoxia gene expression signature in breast tumors from women of diverse ancestry.

EXPERIMENTAL DESIGN

We developed a NanoString nCounter gene expression panel and applied it to breast tumors from Nigeria (n = 182) and the University of Chicago (Chicago, IL; n = 161). We also analyzed RNA sequencing data from Nigeria (n = 84) and The Cancer Genome Atlas (TCGA) datasets (n = 863). Patient prognosis was analyzed using multiple datasets.

RESULTS

The VEGF-hypoxia signature was highest in the basal-like subtype compared with other subtypes, with greater expression in Black women compared with White women. In TCGA dataset, necrotic breast tumors had higher scores for the VEGF-hypoxia signature compared with non-necrosis tumors (P < 0.001), with the highest proportion in the basal-like subtype. Furthermore, necrotic breast tumors have higher scores for the proliferation signature, suggesting an interaction between the VEGF-hypoxia signature, proliferation, and necrosis. T-cell gene expression signatures also correlated with the VEGF-hypoxia signature when testing all tumors in TCGA dataset. Finally, we found a significant association of the VEGF-hypoxia profile with poor outcomes when using all patients in the METABRIC (P < 0.0001) and SCAN-B datasets (P = 0.002).

CONCLUSIONS

These data provide further evidence for breast cancer heterogeneity across diverse populations and molecular subtypes. Interventions selectively targeting VEGF-hypoxia and the immune microenvironment have the potential to improve overall survival in aggressive breast cancers that disproportionately impact Black women in the African Diaspora.

Alleviating misclassified germline variants in underrepresented populations: A strategy using popmax

PURPOSE

Germline variant interpretation often depends on population-matched control cohorts. This is not feasible for population groups that are underrepresented in current population reference databases.

METHODS

We classify germline variants with population-matched controls for 2 ancestrally diverse cohorts of patients: 132 early-onset or familial colorectal carcinoma patients from Singapore and 100 early-onset colorectal carcinoma patients from the United States. The effects of using a population-mismatched control cohort are simulated by swapping the control cohorts used for each patient cohort, with or without the popmax computational strategy.

RESULTS

Population-matched classifications revealed a combined 62 pathogenic or likely pathogenic (P/LP) variants in 34 genes across both cohorts. Using a population-mismatched control cohort resulted in misclassification of non-P/LP variants as P/LP, driven by the absence of ancestry-specific rare variants in the control cohort. Popmax was more effective in alleviating misclassifications for the Singapore cohort than the US cohort.

CONCLUSION

Underrepresented population groups can suffer from higher rates of false-positive P/LP results. Popmax can partially alleviate these misclassifications, but its efficacy still depends on the degree with which the population groups are represented in the control cohort.

Improving access to exome sequencing in a medically underserved population through the Texome Project

PURPOSE

Genomic medicine can end diagnostic odysseys for patients with complex phenotypes; however, limitations in insurance coverage and other systemic barriers preclude individuals from accessing comprehensive genetics evaluation and testing.

METHODS

The Texome Project is a 4-year study that reduces barriers to genomic testing for individuals from underserved and underrepresented populations. Participants with undiagnosed, rare diseases who have financial barriers to obtaining exome sequencing (ES) clinically are enrolled in the Texome Project.

RESULTS

We highlight the Texome Project process and describe the outcomes of the first 60 ES results for study participants. Participants received a genetic evaluation, ES, and return of results at no cost. We summarize the psychosocial or medical implications of these genetic diagnoses. Thus far, ES provided molecular diagnoses for 18 out of 60 (30%) of Texome participants. Plus, in 11 out of 60 (18%) participants, a partial or probable diagnosis was identified. Overall, 5 participants had a change in medical management.

CONCLUSION

To date, the Texome Project has recruited a racially, ethnically, and socioeconomically diverse cohort. The diagnostic rate and medical impact in this cohort support the need for expanded access to genetic testing and services. The Texome Project will continue reducing barriers to genomic care throughout the future study years.

Vitamin D deficiency and VDR gene polymorphism FokI (rs2228570) are associated with diabetes mellitus in adults: COVID-inconfidentes study

BACKGROUND

Diabetes mellitus is a chronic and multifactorial condition, including environmental risk factors such as lifestyle habits and genetic conditions.

OBJECTIVE

We aimed to evaluate the association of VDR gene polymorphism (rs2228570) FokI and vitamin D levels with diabetes in adults.

METHODS

Cross-sectional population-based study in adults, conducted from October to December 2020 in two Brazilian cities. The outcome variable was diabetes, defined as glycated hemoglobin ≥ 6.5% or self-report medical diagnosis or use of oral hypoglycemic drugs. Vitamin D (25-hydroxyvitamin D) was measured by indirect electrochemiluminescence, and classified as deficiency when 25(OH)D < 20 ng/mL. All participants were genotyped for VDR FokI polymorphism by qPCR and classified as homozygous mutant (ff or GG), heterozygous (Ff or AG), or homozygous wild (FF or AA). A combined analysis between the FokI polymorphism and vitamin D levels with diabetes was also examined. A directed acyclic graph (DAG) was used to select minimal and sufficient adjustment for confounding variables by the backdoor criterion.

RESULTS

The prevalence of DM was 9.4% and vitamin D deficiency (VDD) was 19.9%. The genotype distribution of FokI polymorphism was 9.9% FF, 44.8% Ff, and 45.3% ff. It was possible to verify a positive association between vitamin D deficiency and DM (OR = 2.19; 95% CI: 1.06-4.50). Individuals with the altered allele (ff) had a 1.78 higher prevalence of DM (OR: 1.78; 95% CI; 1.10-2.87). Combined analyses, individuals with vitamin D deficiency and one or two copies of the altered FokI allele had a higher prevalence of DM (Ff + ff: OR: 1.67; 95% CI; 1.07-2.61; ff: OR: 3.60; 95% CI; 1.40-9.25).

CONCLUSION

Our data suggest that vitamin D deficiency and FokI polymorphism are associated with DM.

Characterizing substructure via mixture modeling in large-scale genetic summary statistics

Genetic summary data are broadly accessible and highly useful including for risk prediction, causal inference, fine mapping, and incorporation of external controls. However, collapsing individual-level data into groups masks intra- and inter-sample heterogeneity, leading to confounding, reduced power, and bias. Ultimately, unaccounted substructure limits summary data usability, especially for understudied or admixed populations. Here, we present Summix2, a comprehensive set of methods and software based on a computationally efficient mixture model to estimate and adjust for substructure in genetic summary data. In extensive simulations and application to public data, Summix2 characterizes finer-scale population structure, identifies ascertainment bias, and identifies potential regions of selection due to local substructure deviation. Summix2 increases the robust use of diverse publicly available summary data resulting in improved and more equitable research.

Distinct positions of genetic and oral histories: Perspectives from India

Over the past decade, genomic data have contributed to several insights on global human population histories. These studies have been met both with interest and critically, particularly by populations with oral histories that are records of their past and often reference their origins. While several studies have reported concordance between oral and genetic histories, there is potential for tension that may stem from genetic histories being prioritized or used to confirm community-based knowledge and ethnography, especially if they differ. To investigate the interplay between oral and genetic histories, we focused on the southwestern region of India and analyzed whole-genome sequence data from 156 individuals identifying as Bunt, Kodava, Nair, and Kapla. We supplemented limited anthropological records on these populations with oral history accounts from community members and historical literature, focusing on references to non-local origins such as the ancient Scythians in the case of Bunt, Kodava, and Nair, members of Alexander the Great's army for the Kodava, and an African-related source for Kapla. We found these populations to be genetically most similar to other Indian populations, with the Kapla more similar to South Indian tribal populations that maximize a genetic ancestry related to Ancient Ancestral South Indians. We did not find evidence of additional genetic sources in the study populations than those known to have contributed to many other present-day South Asian populations. Our results demonstrate that oral and genetic histories may not always provide consistent accounts of population origins and motivate further community-engaged, multi-disciplinary investigations of non-local origin stories in these communities.

Exploring the relationship between admixture and genetic susceptibility to attention deficit hyperactivity disorder in two Latin American cohorts

Contemporary research on the genomics of Attention Deficit Hyperactivity Disorder (ADHD) often underrepresents admixed populations of diverse genomic ancestries, such as Latin Americans. This study explores the relationship between admixture and genetic associations for ADHD in Colombian and Mexican cohorts. Some 546 participants in two groups, ADHD and Control, were genotyped with Infinium PsychArray®. Global ancestry levels were estimated using overall admixture proportions and principal component analysis, while local ancestry was determined using a method to estimate ancestral components along the genome. Genome-wide association analysis (GWAS) was conducted to identify significant associations. Differences between Colombia and Mexico were evaluated using appropriate statistical tests. 354 Single-nucleotide polymorphisms (SNPs) and Single-nucleotide variants (SNVs) related to some genes and intergenic regions exhibited suggestive significance (p-value < 5*10e−5) in the GWAS. None of the variants revealed genome-wide significance (p-value < 5*10e−8). The study identified a significant relationship between risk SNPs and the European component of admixture, notably observed in the LOC105379109 gene. Despite differences in risk association loci, such as FOXP2, our findings suggest a possible homogeneity in genetic variation’s impact on ADHD between Colombian and Mexican populations. Current reference datasets for ADHD predominantly consist of samples with high European ancestry, underscoring the need for further research to enhance the representation of reference populations and improve the identification of ADHD risk traits in Latin Americans.

Using genome and transcriptome data from African-ancestry female participants to identify putative breast cancer susceptibility genes

African-ancestry (AA) participants are underrepresented in genetics research. Here, we conducted a transcriptome-wide association study (TWAS) in AA female participants to identify putative breast cancer susceptibility genes. We built genetic models to predict levels of gene expression, exon junction, and 3' UTR alternative polyadenylation using genomic and transcriptomic data generated in normal breast tissues from 150 AA participants and then used these models to perform association analyses using genomic data from 18,034 cases and 22,104 controls. At Bonferroni-corrected P < 0.05, we identified six genes associated with breast cancer risk, including four genes not previously reported (CTD-3080P12.3, EN1, LINC01956 and NUP210L). Most of these genes showed a stronger association with risk of estrogen-receptor (ER) negative or triple-negative than ER-positive breast cancer. We also replicated the associations with 29 genes reported in previous TWAS at P < 0.05 (one-sided), providing further support for an association of these genes with breast cancer risk. Our study sheds new light on the genetic basis of breast cancer and highlights the value of conducting research in AA populations.

Imputation accuracy across global human populations

Genotype imputation is now fundamental for genome-wide association studies but lacks fairness due to the underrepresentation of references from non-European ancestries. The state-of-the-art imputation reference panel released by the Trans-Omics for Precision Medicine (TOPMed) initiative improved the imputation of admixed African-ancestry and Hispanic/Latino samples, but imputation for populations primarily residing outside of North America may still fall short in performance due to persisting underrepresentation. To illustrate this point, we imputed the genotypes of over 43,000 individuals across 123 populations around the world and identified numerous populations where imputation accuracy paled in comparison to that of European-ancestry populations. For instance, the mean imputation r-squared (Rsq) for variants with minor allele frequencies between 1% and 5% in Saudi Arabians (n = 1,061), Vietnamese (n = 1,264), Thai (n = 2,435), and Papua New Guineans (n = 776) were 0.79, 0.78, 0.76, and 0.62, respectively, compared to 0.90-0.93 for comparable European populations matched in sample size and SNP array content. Outside of Africa and Latin America, Rsq appeared to decrease as genetic distances to European-ancestry reference increased, as predicted. Using sequencing data as ground truth, we also showed that Rsq may over-estimate imputation accuracy for non-European populations more than European populations, suggesting further disparity in accuracy between populations. Using 1,496 sequenced individuals from Taiwan Biobank as a second reference panel to TOPMed, we also assessed a strategy to improve imputation for non-European populations with meta-imputation, but this design did not improve accuracy across frequency spectra. Taken together, our analyses suggest that we must ultimately strive to increase diversity and size to promote equity within genetics research.

Deciphering the Interplay between Genetic Risk Scores and Lifestyle Factors on Individual Obesity Predisposition

Obesity's variability is significantly influenced by the interplay between genetic and environmental factors. We aimed to integrate the combined impact of genetic risk score (GRSBMI) with physical activity (PA), sugar-sweetened beverages (SSB), wine intake, and eating habits score (EHS) on obesity predisposition risk. Adults' (n = 5824) data were analyzed for common obesity-related single nucleotide polymorphisms and lifestyle habits. The weighted GRSBMI was constructed and categorized into quartiles (Qs), and the adjusted multivariate logistic regression models examined the association of GRSBMI with obesity (BMI ≥ 30) and lifestyle factors. GRSBMI was significantly associated with obesity risk. Each GRSBMI unit was associated with an increase of 3.06 BMI units (p ≤ 0.0001). PA markedly reduced obesity risk across GRSBMI Qs. Inactive participants' (≥90 min/week) mean BMI was higher in GRSBMI Q3-Q4 compared to Q1 (p = 0.003 and p < 0.001, respectively). Scoring EHS ≥ median, SSBs (≥1 cup/day), and non-wine drinking were associated with higher BMI within all GRSBMI Qs compared to EHS < median, non-SSBs, and non-wine drinkers. Mean BMI was higher in GRSBMI Q4 compared to other quartiles (p < 0.0001) in non-wine drinkers and compared to Q1 for SSB's consumers (p = 0.07). A higher GRSBMI augmented the impact of lifestyle factors on obesity. The interplay between GRSBMI and modifiable lifestyle factors provides a tailored personalized prevention and treatment for obesity management.

Research methods at the intersection of gender diversity and autism: A scoping review

LAY ABSTRACT

Research has increasingly focused on the intersection between gender diversity and autism. To better understand this literature, this scoping review systematically searched five databases for peer-reviewed literature on gender diversity and autism published between 2018 and 2023. Included studies (N = 84) were of English language, featured original qualitative or quantitative findings, and examined a psychosocial connection between autism and gender spectra variables. Most studies focused on measuring prevalence of autism among gender-diverse individuals. While the overall study rigor was acceptable, weaknesses in measurement, sample selection, and definition of key terms were noted. Promisingly, studies in this area appear to be shifting away from a pathologizing lens and towards research methods that engage in meaningful collaboration with the autistic, gender-diverse community to investigate how to best enhance the quality of life and wellbeing of this population.

Decoding triancestral origins, archaic introgression, and natural selection in the Japanese population by whole-genome sequencing

We generated Japanese Encyclopedia of Whole-Genome/Exome Sequencing Library (JEWEL), a high-depth whole-genome sequencing dataset comprising 3256 individuals from across Japan. Analysis of JEWEL revealed genetic characteristics of the Japanese population that were not discernible using microarray data. First, rare variant-based analysis revealed an unprecedented fine-scale genetic structure. Together with population genetics analysis, the present-day Japanese can be decomposed into three ancestral components. Second, we identified unreported loss-of-function (LoF) variants and observed that for specific genes, LoF variants appeared to be restricted to a more limited set of transcripts than would be expected by chance, with PTPRD as a notable example. Third, we identified 44 archaic segments linked to complex traits, including a Denisovan-derived segment at NKX6-1 associated with type 2 diabetes. Most of these segments are specific to East Asians. Fourth, we identified candidate genetic loci under recent natural selection. Overall, our work provided insights into genetic characteristics of the Japanese population.

An ensemble penalized regression method for multi-ancestry polygenic risk prediction

Great efforts are being made to develop advanced polygenic risk scores (PRS) to improve the prediction of complex traits and diseases. However, most existing PRS are primarily trained on European ancestry populations, limiting their transferability to non-European populations. In this article, we propose a novel method for generating multi-ancestry Polygenic Risk scOres based on enSemble of PEnalized Regression models (PROSPER). PROSPER integrates genome-wide association studies (GWAS) summary statistics from diverse populations to develop ancestry-specific PRS with improved predictive power for minority populations. The method uses a combination ofL 1 (lasso) andL 2 (ridge) penalty functions, a parsimonious specification of the penalty parameters across populations, and an ensemble step to combine PRS generated across different penalty parameters. We evaluate the performance of PROSPER and other existing methods on large-scale simulated and real datasets, including those from 23andMe Inc., the Global Lipids Genetics Consortium, and All of Us. Results show that PROSPER can substantially improve multi-ancestry polygenic prediction compared to alternative methods across a wide variety of genetic architectures. In real data analyses, for example, PROSPER increased out-of-sample prediction R2 for continuous traits by an average of 70% compared to a state-of-the-art Bayesian method (PRS-CSx) in the African ancestry population. Further, PROSPER is computationally highly scalable for the analysis of large SNP contents and many diverse populations.

An Ensemble Penalized Regression Method for Multi-ancestry Polygenic Risk Prediction

Great efforts are being made to develop advanced polygenic risk scores (PRS) to improve the prediction of complex traits and diseases. However, most existing PRS are primarily trained on European ancestry populations, limiting their transferability to non-European populations. In this article, we propose a novel method for generating multi-ancestry Polygenic Risk scOres based on enSemble of PEnalized Regression models (PROSPER). PROSPER integrates genome-wide association studies (GWAS) summary statistics from diverse populations to develop ancestry-specific PRS with improved predictive power for minority populations. The method uses a combination of ℒ 1 (lasso) and ℒ 2 (ridge) penalty functions, a parsimonious specification of the penalty parameters across populations, and an ensemble step to combine PRS generated across different penalty parameters. We evaluate the performance of PROSPER and other existing methods on large-scale simulated and real datasets, including those from 23andMe Inc., the Global Lipids Genetics Consortium, and All of Us. Results show that PROSPER can substantially improve multi-ancestry polygenic prediction compared to alternative methods across a wide variety of genetic architectures. In real data analyses, for example, PROSPER increased out-of-sample prediction R2 for continuous traits by an average of 70% compared to a state-of-the-art Bayesian method (PRS-CSx) in the African ancestry population. Further, PROSPER is computationally highly scalable for the analysis of large SNP contents and many diverse populations.

Ancestry-specific high-risk gene variant profiling unmasks diabetes-associated genes

How ancestry-associated genetic variance affects disparities in the risk of polygenic diseases and influences the identification of disease-associated genes warrants a deeper understanding. We hypothesized that the discovery of genes associated with polygenic diseases may be limited by the overreliance on single-nucleotide polymorphism (SNP)-based genomic investigation, as most significant variants identified in genome-wide SNP association studies map to introns and intergenic regions of the genome. To overcome such potential limitations, we developed a gene-constrained, function-based analytical method centered on high-risk variants (hrV) that encode frameshifts, stopgains or splice site disruption. We analyzed the total number of hrV per gene in populations of different ancestry, representing a total of 185 934 subjects. Using this analysis, we developed a quantitative index of hrV (hrVI) across 20 428 genes within each population. We then applied hrVI analysis to the discovery of genes associated with type 2 diabetes mellitus (T2DM), a polygenic disease with ancestry-related disparity. HrVI profiling and gene-to-gene comparisons of ancestry-specific hrV between the case (20 781 subjects) and control (24 440 subjects) populations in the T2DM national repository identified 57 genes associated with T2DM, 40 of which were discoverable only by ancestry-specific analysis. These results illustrate how a function-based, ancestry-specific analysis of genetic variations can accelerate the identification of genes associated with polygenic diseases. Besides T2DM, such analysis may facilitate our understanding of the genetic basis for other polygenic diseases that are also greatly influenced by environmental and behavioral factors, such as obesity, hypertension and Alzheimer's disease.

Immunogenetic Diversity and Cancer Immunotherapy Disparities

SUMMARY

The success of checkpoint blockade cancer immunotherapies has unequivocally confirmed the critical role of T cells in cancer immunity and boosted the development of immunotherapeutic strategies targeting specific antigens on cancer cells. The vast immunogenetic diversity of human leukocyte antigen (HLA) class I alleles across populations is a key factor influencing the advancement of HLA class I-restricted therapies and related research and diagnostic tools.

Adjusting for principal components can induce spurious associations in genome-wide association studies in admixed populations

Principal component analysis (PCA) is widely used to control for population structure in genome-wide association studies (GWAS). Top principal components (PCs) typically reflect population structure, but challenges arise in deciding how many PCs are needed and ensuring that PCs do not capture other artifacts such as regions with atypical linkage disequilibrium (LD). In response to the latter, many groups suggest performing LD pruning or excluding known high LD regions prior to PCA. However, these suggestions are not universally implemented and the implications for GWAS are not fully understood, especially in the context of admixed populations. In this paper, we investigate the impact of pre-processing and the number of PCs included in GWAS models in African American samples from the Women's Women's Health Initiative SNP Health Association Resource and two Trans-Omics for Precision Medicine Whole Genome Sequencing Project contributing studies (Jackson Heart Study and Genetic Epidemiology of Chronic Obstructive Pulmonary Disease Study). In all three samples, we find the first PC is highly correlated with genome-wide ancestry whereas later PCs often capture local genomic features. The pattern of which, and how many, genetic variants are highly correlated with individual PCs differs from what has been observed in prior studies focused on European populations and leads to distinct downstream consequences: adjusting for such PCs yields biased effect size estimates and elevated rates of spurious associations due to the phenomenon of collider bias. Excluding high LD regions identified in previous studies does not resolve these issues. LD pruning proves more effective, but the optimal choice of thresholds varies across datasets. Altogether, our work highlights unique issues that arise when using PCA to control for ancestral heterogeneity in admixed populations and demonstrates the importance of careful pre-processing and diagnostics to ensure that PCs capturing multiple local genomic features are not included in GWAS models.

Overcoming barriers to single-cell RNA sequencing adoption in low- and middle-income countries

The advent of single-cell resolution sequencing and spatial transcriptomics has enabled the delivery of cellular and molecular atlases of tissues and organs, providing new insights into tissue health and disease. However, if the full potential of these technologies is to be equitably realised, ancestrally inclusivity is paramount. Such a goal requires greater inclusion of both researchers and donors in low- and middle-income countries (LMICs). In this perspective, we describe the current landscape of ancestral inclusivity in genomic and single-cell transcriptomic studies. We discuss the collaborative efforts needed to scale the barriers to establishing, expanding, and adopting single-cell sequencing research in LMICs and to enable globally impactful outcomes of these technologies.

Genetic Researchers' Use of and Interest in Research With Diverse Ancestral Groups

Importance

Genetic researchers must have access to databases populated with data from diverse ancestral groups to ensure research is generalizable or targeted for historically excluded communities.

Objective

To determine genetic researchers' interest in doing research with diverse ancestral populations, which database stewards offer adequate samples, and additional facilitators for use of diverse ancestral data.

Design, Setting, and Participants

This survey study was conducted from June to December 2022 and was part of an exploratory sequential mixed-methods project in which previous qualitative results informed survey design. Eligible participants included genetic researchers who held US academic affiliations and conducted research using human genetic databases.

Exposure

Internet-administered survey to genetic research professionals.

Main Outcomes and Measures

The survey assessed respondents' experience and interest in research with diverse ancestral data, perceptions of adequacy of diverse data across database stewards (ie, private, government, or consortia), and identified facilitators for encouraging use of diverse ancestral data. Descriptive statistics, χ2 tests, and z tests were used to describe respondents' perspectives and experiences.

Results

A total of 294 researchers (171 men [58.5%]; 121 women [41.2%]) were included in the study, resulting in a response rate of 20.4%. Across seniority level, 109 respondents (37.1%) were senior researchers, 85 (28.9%) were mid-level researchers, 71 (24.1%) were junior researchers, and 27 (9.2%) were trainees. Significantly more respondents worked with data from European ancestral populations (261 respondents [88.8%]) compared with any other ancestral population. Respondents who had not done research with Indigenous ancestral groups (210 respondents [71.4%]) were significantly more likely to report interest in doing so than not (121 respondents [41.2%] vs 89 respondents [30.3%]; P < .001). Respondents reported discrepancies in the adequacy of ancestral populations with significantly more reporting European samples as adequate across consortium (203 respondents [90.6%]), government (200 respondents [89.7%]), and private (42 respondents [80.8%]) databases, compared with any other ancestral population. There were no significant differences in reported adequacy of ancestral populations across database stewards. A majority of respondents without access to adequate diverse samples reported that increasing the ancestral diversity of existing databases (201 respondents [68.4%]) and increasing access to databases that are already diverse (166 respondents [56.5%]) would increase the likelihood of them using a more diverse sample.

Conclusions and Relevance

In this survey study of US genetic researchers, respondents reported existing databases only provide adequate ancestral samples for European populations, despite their interest in other ancestral populations. These findings suggest there are specific gaps in access to and composition of genetic databases, highlighting the urgent need to boost diversity in research samples to improve inclusivity in genetic research practices.

Molecular testing in newborn screening: VUS burden among true positives and secondary reproductive limitations via expanded carrier screening panels

PURPOSE

Expanded carrier screening (ECS) gene panels have several limitations, including variable content, current knowledge of disease-causing variants, and differing reporting policies. This study evaluated if the disease-associated variants identified in affected neonates who screened positive by California newborn screening (NBS) for an inherited metabolic disorder (IMD) by tandem mass spectrometry (MS/MS) would likely be reported by ECS gene panels.

METHODS

Retrospective review of neonates referred by the California Department of Public Health for a positive NBS by multianalyte MS/MS from January 1, 2020 through June 30, 2021.

RESULTS

One hundred thirty-six neonates screened positive for ≥1 NBS MS/MS indication. Nineteen neonates (14%) were ultimately diagnosed with an IMD, all of whom had abnormal biochemical testing. Eighteen of the 19 underwent molecular testing; 10 (56%) neonates had ≥1 variants of uncertain significance, 9 of whom were of non-White ancestry. ECS panels would have been negative for 56% (20/36) of parents with an affected neonate, 85% (17/20) of whom were of non-White ancestry.

CONCLUSION

The number of variants of uncertain significance identified in this cohort highlights the need for more diversified variant databases. Due in part to the lack of diversity in currently sequenced populations, genomic sequencing cannot replace biochemical testing for the diagnosis of an IMD.

Genetic control of DNA methylation is largely shared across European and East Asian populations

DNA methylation is an ideal trait to study the extent of the shared genetic control across ancestries, effectively providing hundreds of thousands of model molecular traits with large QTL effect sizes. We investigate cis DNAm QTLs in three European (n = 3701) and two East Asian (n = 2099) cohorts to quantify the similarities and differences in the genetic architecture across populations. We observe 80,394 associated mQTLs (62.2% of DNAm probes with significant mQTL) to be significant in both ancestries, while 28,925 mQTLs (22.4%) are identified in only a single ancestry. mQTL effect sizes are highly conserved across populations, with differences in mQTL discovery likely due to differences in allele frequency of associated variants and differing linkage disequilibrium between causal variants and assayed SNPs. This study highlights the overall similarity of genetic control across ancestries and the value of ancestral diversity in increasing the power to detect associations and enhancing fine mapping resolution.

Key Contextual Factors Involved with Participation in Medical and Genomic Screening and Research for African American and Caucasian Americans: A Qualitative Inquiry American Journal of Community Genetics

Tremendous progress has been made promoting diversity in recruitment for genomic research, yet challenges remain for several racial demographics. Research has cited intertwined fears of racial discrimination and medical mistrust as contributing factors. This study aimed to identify key factors to establishing trust in medical and genomic screening and research among African Americans and White Americans. Participants completed online focus groups and resulting transcripts were analyzed using a qualitative descriptive approach, with content analysis methods based on recommendations by Schreier. Fifteen African Americans and 23 Caucasian Americans participated in the study, 63% of which were female. The mean age of participants was 38.53 (SD = 16.6). The Overarching Theme of Trust is Context Dependent was identified, along with the following five themes describing elements influencing trustworthiness for our participants: 1) Professional Experience, Education, and Training Bolster Trust; 2) Trust Depends on Relationships; 3) Cross-checking Provided Information is Influential in Establishing Trust; 4) Trust is Undermined by Lack of Objectivity and Bias; and 5) Racism is an Embedded Concern and a Medical Trust Limiting Component for African Americans. To effectively address mistrust and promote recruitment of diverse participants, genomic research initiatives must be communicated in a manner that resonates with the specific diverse communities targeted. Our results suggest key factors influencing trust that should be attended to if we are to promote equity appropriately and respectfully by engaging diverse populations in genomic research.

Ancestry, race and ethnicity: the role and relevance of language in clinical genetics practice

BACKGROUND

The terms ancestry, race and ethnicity are used variably within the medical literature and within society and clinical care. Biological lineage can provide an important context for the interpretation of genomic data, but the language used, and practices around when to ascertain this, vary.

METHODS

Using a fictional case scenario we explore the relevance of questions around ancestry, race and ethnicity in clinical genetic practice.

RESULTS

In the UK, data on 'ethnicity' are routinely collected by those using genomic medicine, as well as within the wider UK National Health Service, although the reasons for this are not always clear to practitioners and patients. Sometimes it is requested as a proxy for biological lineage to aid variant interpretation, refine estimations of carrier frequency and guide decisions around the need for pharmacogenetic testing.

CONCLUSION

There are many challenges around the use and utility of these terms. Currently, genomic databases are populated primarily with data from people of European descent, and this can lead to health disparities and poorer service for minoritised or underserved populations. Sensitivity and consideration are needed when communicating with patients around these areas. We explore the role and relevance of language around biological lineage in clinical genetics practice.

Population-enriched innate immune variants may identify candidate gene targets at the intersection of cancer and cardio-metabolic disease

Both cancer and cardio-metabolic disease disparities exist among specific populations in the US. For example, African Americans experience the highest rates of breast and prostate cancer mortality and the highest incidence of obesity. Native and Hispanic Americans experience the highest rates of liver cancer mortality. At the same time, Pacific Islanders have the highest death rate attributed to type 2 diabetes (T2D), and Asian Americans experience the highest incidence of non-alcoholic fatty liver disease (NAFLD) and cancers induced by infectious agents. Notably, the pathologic progression of both cancer and cardio-metabolic diseases involves innate immunity and mechanisms of inflammation. Innate immunity in individuals is established through genetic inheritance and external stimuli to respond to environmental threats and stresses such as pathogen exposure. Further, individual genomes contain characteristic genetic markers associated with one or more geographic ancestries (ethnic groups), including protective innate immune genetic programming optimized for survival in their corresponding ancestral environment(s). This perspective explores evidence related to our working hypothesis that genetic variations in innate immune genes, particularly those that are commonly found but unevenly distributed between populations, are associated with disparities between populations in both cancer and cardio-metabolic diseases. Identifying conventional and unconventional innate immune genes that fit this profile may provide critical insights into the underlying mechanisms that connect these two families of complex diseases and offer novel targets for precision-based treatment of cancer and/or cardio-metabolic disease.

Population Neuroscience: Strategies to Promote Data Sharing While Protecting Privacy

Population neuroscience aims to advance our understanding of how genetic and environmental factors influence brain development and brain health over the life span, by integrating genomics, epidemiology, and neuroscience at population scale. This big data approach depends on data sharing strategies at both the micro- and macro-level, as well as attention to effective data management and protection of participant privacy. At the micro-level, researchers participate in international consortia that support collaboration, standards, and data sharing. They also seek to link together cohort studies, administrative health databases, and measures of the physical, built, and social environment in creative ways. Large-scale, longitudinal, and multi-modal cohorts are being designed to support explorations of genetic and environmental impacts on the brain. At a macro-level, funding agency policies now require data across health research domains to be managed according to the FAIR (findable, accessible, interoperable, and re-useable) Data principles and made available to the research community in a timely manner to support reproducibility and re-use. Data repositories provide technical infrastructure for storing, accessing, and increasingly also analyzing rich population-level data. Federated and cloud-based approaches are being leveraged to improve the security, remote accessibility, and performance of repositories. Finally, legal frameworks are being developed to facilitate secure health data access, integration, and analysis, providing new opportunities for the field.

Genotype prediction of 336,463 samples from public expression data

Tens of thousands of RNA-sequencing experiments comprising hundreds of thousands of individual samples have now been performed. These data represent a broad range of experimental conditions, sequencing technologies, and hypotheses under study. The Recount project has aggregated and uniformly processed hundreds of thousands of publicly available RNA-seq samples. Most of these samples only include RNA expression measurements; genotype data for these same samples would enable a wide range of analyses including variant prioritization, eQTL analysis, and studies of allele specific expression. Here, we developed a statistical model based on the existing reference and alternative read counts from the RNA-seq experiments available through Recount3 to predict genotypes at autosomal biallelic loci in coding regions. We demonstrate the accuracy of our model using large-scale studies that measured both gene expression and genotype genome-wide. We show that our predictive model is highly accurate with 99.5% overall accuracy, 99.6% major allele accuracy, and 90.4% minor allele accuracy. Our model is robust to tissue and study effects, provided the coverage is high enough. We applied this model to genotype all the samples in Recount 3 and provide the largest ready-to-use expression repository containing genotype information. We illustrate that the predicted genotype from RNA-seq data is sufficient to unravel the underlying population structure of samples in Recount3 using Principal Component Analysis.

Impact of AI-Powered Solutions in Rehabilitation Process: Recent Improvements and Future Trends

Rehabilitation is an important and necessary part of local and global healthcare services along with treatment and palliative care, prevention of disease, and promotion of good health. The rehabilitation process helps older and young adults even children to become as independent as possible in activities of daily life and enables participation in useful living activities, recreation, work, and education. The technology of Artificial Intelligence (AI) has evolved significantly in recent years. Many activities related to rehabilitation have been getting benefits from using AI techniques. The objective of this review study is to explore the advantages of AI for rehabilitation and how AI is impacting the rehabilitation process. This study aims at the most critical aspects of the rehabilitation process that could potentially take advantage of AI techniques including personalized rehabilitation apps, rehabilitation through assistance, rehabilitation for neurological disorders, rehabilitation for developmental disorders, virtual reality rehabilitation, rehabilitation of neurodegenerative diseases and Telerehabilitation of Cardiovascular. We presented a survey on the newest empirical studies available in the literature including the AI-based technology helpful in the Rehabilitation process. The novelty feature included but was not limited to an overview of the technological solutions useful in rehabilitation. Seven different categories were identified. Illustrative examples of practical applications were detailed. Implications of the findings for both research and practice were critically discussed. Most of the AI applications in these rehabilitation types are in their infancy and continue to grow while exploring new opportunities. Therefore, we investigate the role of AI technology in rehabilitation processes. In addition, we do statistical analysis of the selected studies to highlight the significance of this review work. In the end, we also present a discussion on some challenges, and future research directions.

The association of GNB5 with Alzheimer disease revealed by genomic analysis restricted to variants impacting gene function

Disease-associated variants identified from genome-wide association studies (GWASs) frequently map to non-coding areas of the genome such as introns and intergenic regions. An exclusive reliance on gene-agnostic methods of genomic investigation could limit the identification of relevant genes associated with polygenic diseases such as Alzheimer disease (AD). To overcome such potential restriction, we developed a gene-constrained analytical method that considers only moderate- and high-risk variants that affect gene coding sequences. We report here the application of this approach to publicly available datasets containing 181,388 individuals without and with AD and the resulting identification of 660 genes potentially linked to the higher AD prevalence among Africans/African Americans. By integration with transcriptome analysis of 23 brain regions from 2,728 AD case-control samples, we concentrated on nine genes that potentially enhance the risk of AD: AACS, GNB5, GNS, HIPK3, MED13, SHC2, SLC22A5, VPS35, and ZNF398. GNB5, the fifth member of the heterotrimeric G protein beta family encoding Gβ5, is primarily expressed in neurons and is essential for normal neuronal development in mouse brain. Homozygous or compound heterozygous loss of function of GNB5 in humans has previously been associated with a syndrome of developmental delay, cognitive impairment, and cardiac arrhythmia. In validation experiments, we confirmed that Gnb5 heterozygosity enhanced the formation of both amyloid plaques and neurofibrillary tangles in the brains of AD model mice. These results suggest that gene-constrained analysis can complement the power of GWASs in the identification of AD-associated genes and may be more broadly applicable to other polygenic diseases.

GUANinE v1.0: Benchmark Datasets for Genomic AI Sequence-to-Function Models

Computational genomics increasingly relies on machine learning methods for genome interpretation, and the recent adoption of neural sequence-to-function models highlights the need for rigorous model specification and controlled evaluation, problems familiar to other fields of AI. Research strategies that have greatly benefited other fields - including benchmarking, auditing, and algorithmic fairness - are also needed to advance the field of genomic AI and to facilitate model development. Here we propose a genomic AI benchmark, GUANinE, for evaluating model generalization across a number of distinct genomic tasks. Compared to existing task formulations in computational genomics, GUANinE is large-scale, de-noised, and suitable for evaluating pretrained models. GUANinE v1.0 primarily focuses on functional genomics tasks such as functional element annotation and gene expression prediction, and it also draws upon connections to evolutionary biology through sequence conservation tasks. The current GUANinE tasks provide insight into the performance of existing genomic AI models and non-neural baselines, with opportunities to be refined, revisited, and broadened as the field matures. Finally, the GUANinE benchmark allows us to evaluate new self-supervised T5 models and explore the tradeoffs between tokenization and model performance, while showcasing the potential for self-supervision to complement existing pretraining procedures.

Tackling reference bias in genotyping by using founder sequences with PanVC 3

Summary

Overcoming reference bias and calling insertions and deletions are major challenges in genotyping. We present PanVC 3, a set of software that can be utilized as part of various variant calling workflows. We show that, by incorporating known genetic variants to a set of founder sequences to which reads are aligned, reference bias is reduced and precision of calling insertions and deletions is improved.

Availability and implementation

PanVC 3 and its source code are freely available at https://github.com/tsnorri/panvc3 and at https://anaconda.org/tsnorri/panvc3 under the MIT licence. The experiment scripts are available at https://github.com/algbio/panvc3-experiments.