Alzheimer's Disease and Related Dementias (ADRD) afflict almost 7 million people in the USA alone. The majority of research in ADRD is conducted using post-mortem samples of brain tissue or carefully recruited clinical trial patients. While these resources are excellent, they suffer from lack of sex/gender, and racial/ethnic inclusiveness. Electronic Health Records (EHR) data has the potential to bridge this gap by including real-world ADRD patients treated during routine clinical care. In this study, we utilize EHR data from a cohort of 70,420 ADRD patients diagnosed and treated at Penn Medicine. Our goal is to uncover important risk features leading to three types of Neuro-Degenerative Disorders (NDD), including Alzheimer's Disease (AD), Parkinson's Disease (PD) and Other Dementias (OD). We employ a variety of Machine Learning (ML) Methods, including uni-variate and multivariate ML approaches and compare accuracies across the ML methods. We also investigate the types of features identified by each method, the overlapping features and the unique features to highlight important advantages and disadvantages of each approach specific for certain NDD types. Our study is important for those interested in studying ADRD and NDD in EHRs as it highlights the strengths and limitations of popular approaches employed in the ML community. We found that the uni-variate approach was able to uncover features that were important and rare for specific types of NDD (AD, PD, OD), which is important from a clinical perspective. Features that were found across all methods represent features that are the most robust.

Large language models (LLMs) and biomedical annotations have a symbiotic relationship. LLMs rely on high-quality annotations for training and/or fine-tuning for specific biomedical tasks. These annotations are traditionally generated through expensive and time-consuming human curation. Meanwhile LLMs can also be used to accelerate the process of curation, thus simplifying the process, and potentially creating a virtuous feedback loop. However, their use also introduces new limitations and risks, which are as important to consider as the opportunities they offer. In this workshop, we will review the process that has led to the current rise of LLMs in several fields, and in particular in biomedicine, and discuss specifically the opportunities and pitfalls when they are applied to biomedical annotation and curation.

Wearable silicone wristbands are a rapidly growing exposure assessment technology that offer researchers the ability to study previously inaccessible cohorts and have the potential to provide a more comprehensive picture of chemical exposure within diverse communities. However, there are no established best practices for analyzing the data within a study or across multiple studies, thereby limiting impact and access of these data for larger meta-analyses. We utilize data from three studies, from over 600 wristbands worn by participants in New York City and Eugene, Oregon, to present a first-of-its-kind manuscript detailing wristband data properties. We further discuss and provide concrete examples of key areas and considerations in common statistical modeling methods where best practices must be established to enable meta-analyses and integration of data from multiple studies. Finally, we detail important and challenging aspects of machine learning, meta-analysis, and data integration that researchers will face in order to extend beyond the limited scope of individual studies focused on specific populations.

Cells consist of large components, such as organelles, that recursively factor into smaller systems, such as condensates and protein complexes, forming a dynamic multi-scale structure of the cell. Recent technological innovations have paved the way for systematic interrogation of subcellular structures, yielding unprecedented insights into their roles and interactions. In this workshop, we discuss progress, challenges, and collaboration to marshal various computational approaches toward assembling an integrated structural map of the human cell.

Existing proposed pathogenesis for preeclampsia (PE) was only applied for early onset subtype and did not consider pre-pregnancy and competing risks. We aimed to decipher PE subtypes by identifying related transcriptome that represents endometrial maturation and histologic chorioamnionitis. We utilized eight arrays of mRNA expression for discovery (n=289), and other eight arrays for validation (n=352). Differentially expressed genes (DEGs) were overlapped between those of: (1) healthy samples from endometrium, decidua, and placenta, and placenta samples under histologic chorioamnionitis; and (2) placenta samples for each of the subtypes. They were all possible combinations based on four axes: (1) pregnancy-induced hypertension; (2) placental dysfunction-related diseases (e.g., fetal growth restriction [FGR]); (3) onset; and (4) severity. The DEGs of endometrium at late-secretory phase, but none of decidua, significantly overlapped with those of any subtypes with: (1) early onset (p-values ≤0.008); (2) severe hypertension and proteinuria (p-values ≤0.042); or (3) chronic hypertension and/or severe PE with FGR (p-values ≤0.042). Although sharing the same subtypes whose DEGs with which significantly overlap, the gene regulation was mostly counter-expressed in placenta under chorioamnionitis (n=13/18, 72.22%; odds ratio [OR] upper bounds ≤0.21) but co-expressed in late-secretory endometrium (n=3/9, 66.67%; OR lower bounds ≥1.17). Neither the placental DEGs at first-nor second-trimester under normotensive pregnancy significantly overlapped with those under late-onset, severe PE without FGR. We identified the transcriptome of endometrial maturation in placental dysfunction that distinguished early- and late-onset PE, and indicated chorioamnionitis as a PE competing risk. This study implied a feasibility to develop and validate the pathogenesis models that include pre-pregnancy and competing risks to decide if it is needed to collect prospective data for PE starting from pre-pregnancy including chorioamnionitis information.

Subcellular protein localization is important for understanding functional states of cells, but measuring and quantifying this information can be difficult and typically requires high-resolution microscopy. In this work, we develop a metric to define surface protein polarity from immunofluorescence (IF) imaging data and use it to identify distinct immune cell states within tumor microenvironments. We apply this metric to characterize over two million cells across 600 patient samples and find that cells identified as having polar expression exhibit characteristics relating to tumor-immune cell engagement. Additionally, we show that incorporating these polarity-defined cell subtypes improves the performance of deep learning models trained to predict patient survival outcomes. This method provides a first look at using subcellular protein expression patterns to phenotype immune cell functional states with applications to precision medicine.

The lack of diversity in genomic datasets, currently skewed towards individuals of European ancestry, presents a challenge in developing inclusive biomedical models. The scarcity of such data is particularly evident in labeled datasets that include genomic data linked to electronic health records. To address this gap, this paper presents PopGenAdapt, a genotype-to-phenotype prediction model which adopts semi-supervised domain adaptation (SSDA) techniques originally proposed for computer vision. PopGenAdapt is designed to leverage the substantial labeled data available from individuals of European ancestry, as well as the limited labeled and the larger amount of unlabeled data from currently underrepresented populations. The method is evaluated in underrepresented populations from Nigeria, Sri Lanka, and Hawaii for the prediction of several disease outcomes. The results suggest a significant improvement in the performance of genotype-to-phenotype models for these populations over state-of-the-art supervised learning methods, setting SSDA as a promising strategy for creating more inclusive machine learning models in biomedical research.Our code is available at https://github.com/AI-sandbox/PopGenAdapt.

As the last decade of human genomics research begins to bear the fruit of advancements in precision medicine, it is important to ensure that genomics’ improvements in human health are distributed globally and equitably. An important step to ensuring health equity is to improve the human reference genome to capture global diversity by including a wide variety of alternative haplotypes, sequences that are not currently captured on the reference genome. We present a method that localizes 100 basepair (bp) long sequences extracted from short-read sequencing that can ultimately be used to identify what regions of the human genome non-reference sequences belong to. We extract reads that don’t align to the reference genome, and compute the population’s distribution of 100-mers found within the unmapped reads. We use genetic data from families to identify shared genetic material between siblings and match the distribution of unmapped k-mers to these inheritance patterns to determine the the most likely genomic region of a k-mer. We perform this localization with two highly interpretable methods of artificial intelligence: a computationally tractable Hidden Markov Model coupled to a Maximum Likelihood Estimator. Using a set of alternative haplotypes with known locations on the genome, we show that our algorithm is able to localize 96% of k-mers with over 90% accuracy and less than 1Mb median resolution. As the collection of sequenced human genomes grows larger and more diverse, we hope that this method can be used to improve the human reference genome, a critical step in addressing precision medicine’s diversity crisis.

Big Data neuroimaging collaborations including Enhancing Neuro Imaging Genetics through Meta-Analysis (ENIGMA) integrated worldwide data to identify regional brain deficits in major depressive disorder (MDD). We evaluated the sensitivity of translating ENIGMA-defined MDD deficit patterns to the individual level. We treated ENIGMA MDD deficit patterns as a vector to gauge the similarity between individual and MDD patterns by calculating ENIGMA dot product (EDP). We analyzed the sensitivity and specificity of EDP in separating subjects with (1) subclinical depressive symptoms without a diagnosis of MDD, (2) single episode MDD, (3) recurrent MDD, and (4) controls free of neuropsychiatric disorders. We compared EDP to the Quantile Regression Index (QRI; a linear alternative to the brain age metric) and the global gray matter thickness and subcortical volumes and fractional anisotropy (FA) of water diffusion. We performed this analysis in a large epidemiological sample of UK Biobank (UKBB) participants (N=17,053/19,265 M/F). Group-average increases in depressive symptoms from controls to recurrent MDD was mirrored by EDP (r2=0.85), followed by FA (r2=0.81) and QRI (r2=0.56). Subjects with MDD showed worse performance on cognitive tests than controls with deficits observed for 3 out of 9 cognitive tests administered by the UKBB. We calculated correlations of EDP and other brain indices with measures of cognitive performance in controls. The correlation pattern between EDP and cognition in controls was similar (r2=0.75) to the pattern of cognitive differences in MDD. This suggests that the elevation in EDP, even in controls, is associated with cognitive performance - specifically in the MDD-affected domains. That specificity was missing for QRI, FA or other brain imaging indices. In summary, translating anatomically informed meta-analytic indices of similarity using a linear vector approach led to better sensitivity to depressive symptoms and cognitive patterns than whole-brain imaging measurements or an index of accelerated aging.

Intimate partner violence (IPV) is an important social and public health problem, affecting millions of women worldwide. Violence in a relationship can occur in multiple ways, including physical violence, psychological aggression, and sexual violence. In this study, utilizing data from the National Intimate Partner and Sexual Violence Survey (NISVS), we comprehensively investigate the interplay between physical, psychological, and sexual violence, in terms of their co-occurrence patterns, their relation to trauma symptoms and overall health of victims. For this purpose, we perform network analysis and develop a visualization technique that enables in-depth navigation of the three-dimensional (physical, psychological, sexual) space of violence. Our findings show that physical violence tends to significantly co-occur with psychological abuse, and violence intensifies when both are present. We also find that sexual violence tends to overlap less with other types of violence, particularly with physical violence. Milder forms of psychological abuse are prominent in the population and seem to represent a separate type of abuse (micro-aggression) in terms of its occurrence patterns. Finally, we observe that trauma symptoms and health problems tend to be reported more by survivors at the presence of intense psychological aggression. Our findings can be useful in developing treatments that target different patterns of IPV.