Pacific Symposium on Biocomputing. Pacific Symposium on Biocomputing最新文献

Classical machine learning and deep learning models for Computer-Aided Diagnosis (CAD) commonly focus on overall classification performance, treating misclassification errors (false negatives and false positives) equally during training. This uniform treatment overlooks the distinct costs associated with each type of error, leading to suboptimal decision-making, particularly in the medical domain where it is important to improve the prediction sensitivity without significantly compromising overall accuracy. This study introduces a novel deep learning-based CAD system that incorporates a cost-sensitive parameter into the activation function. By applying our methodologies to two medical imaging datasets, our proposed study shows statistically significant increases of 3.84% and 5.4% in sensitivity while maintaining overall accuracy for Lung Image Database Consortium (LIDC) and Breast Cancer Histological Database (BreakHis), respectively. Our findings underscore the significance of integrating cost-sensitive parameters into future CAD systems to optimize performance and ultimately reduce costs and improve patient outcomes.

This study quantifies health outcome disparities in invasive Methicillin-Resistant Staphylococcus aureus (MRSA) infections by leveraging a novel artificial intelligence (AI) fairness algorithm, the Fairness-Aware Causal paThs (FACTS) decomposition, and applying it to real-world electronic health record (EHR) data. We spatiotemporally linked 9 years of EHRs from a large healthcare provider in Florida, USA, with contextual social determinants of health (SDoH). We first created a causal structure graph connecting SDoH with individual clinical measurements before/upon diagnosis of invasive MRSA infection, treatments, side effects, and outcomes; then, we applied FACTS to quantify outcome potential disparities of different causal pathways including SDoH, clinical and demographic variables. We found moderate disparity with respect to demographics and SDoH, and all the top ranked pathways that led to outcome disparities in age, gender, race, and income, included comorbidity. Prior kidney impairment, vancomycin use, and timing were associated with racial disparity, while income, rurality, and available healthcare facilities contributed to gender disparity. From an intervention standpoint, our results highlight the necessity of devising policies that consider both clinical factors and SDoH. In conclusion, this work demonstrates a practical utility of fairness AI methods in public health settings.

The problem of microdissection of heterogeneous tissue samples is of great interest for both fundamental biology and biomedical research. Until now, microdissection in the form of supervised deconvolution of mixed sequencing samples has been limited to assays measuring gene expression (RNA-seq) or chromatin accessibility (ATAC-seq). We present here the first attempt at solving the supervised deconvolution problem for run-on nascent sequencing data (GRO-seq and PRO-seq), a readout of active transcription. Then, we develop a novel filtering method suited to the mixed set of promoter and enhancer regions provided by nascent sequencing, and apply best-practice standards from the RNA-seq literature, using in-silico mixtures of cells. Using these methods, we find that enhancer RNAs are highly informative features for supervised deconvolution. In most cases, simple deconvolution methods perform better than more complex ones for solving the nascent deconvolution problem. Furthermore, undifferentiated cell types confound deconvolution of nascent sequencing data, likely as a consequence of transcriptional activity over the highly open chromatin regions of undifferentiated cell types. Our results suggest that while the problem of nascent deconvolution is generally tractable, stronger approaches integrating other sequencing protocols may be required to solve mixtures containing undifferentiated celltypes.

Protein-protein interactions play an essential role in nearly all biological processes, and it has become increasingly clear that in order to better understand the fundamental processes that underlie disease, we must develop a strong understanding of both their context specificity (e.g., tissue-specificity) as well as their dynamic nature (e.g., how they respond to environmental changes). While network-based approaches have found much initial success in the application of protein-protein interactions (PPIs) towards systems-level explorations of biology, they often overlook the fact that large numbers of proteins undergo alternative splicing. Alternative splicing has not only been shown to diversify protein function through the generation of multiple protein isoforms, but also remodel PPIs and affect a wide range diseases, including cancer. Isoform-specific interactions are not well characterized, so we develop a computational approach that uses domain-domain interactions in concert with differential exon usage data from The Cancer Genome Atlas (TCGA) and the Genotype-Tissue Expression project (GTEx). Using this approach, we can characterize PPIs likely disrupted or possibly even increased due to splicing events for individual TCGA cancer patient samples relative to a matched GTEx normal tissue background.

This work demonstrates the use of cluster analysis in detecting fair and unbiased novel discoveries. Given a sample population of elective spinal fusion patients, we identify two overarching subgroups driven by insurance type. The Medicare group, associated with lower socioeconomic status, exhibited an over-representation of negative risk factors. The findings provide a compelling depiction of the interwoven socioeconomic and racial disparities present within the healthcare system, highlighting their consequential effects on health inequalities. The results are intended to guide design of fair and precise machine learning models based on intentional integration of population stratification.

Polygenic risk scores (PRS) have predominantly been derived from genome-wide association studies (GWAS) conducted in European ancestry (EUR) individuals. In this study, we present an in-depth evaluation of PRS based on multi-ancestry GWAS for five cardiometabolic phenotypes in the Penn Medicine BioBank (PMBB) followed by a phenome-wide association study (PheWAS). We examine the PRS performance across all individuals and separately in African ancestry (AFR) and EUR ancestry groups. For AFR individuals, PRS derived using the multi-ancestry LD panel showed a higher effect size for four out of five PRSs (DBP, SBP, T2D, and BMI) than those derived from the AFR LD panel. In contrast, for EUR individuals, the multi-ancestry LD panel PRS demonstrated a higher effect size for two out of five PRSs (SBP and T2D) compared to the EUR LD panel. These findings underscore the potential benefits of utilizing a multi-ancestry LD panel for PRS derivation in diverse genetic backgrounds and demonstrate overall robustness in all individuals. Our results also revealed significant associations between PRS and various phenotypic categories. For instance, CAD PRS was linked with 18 phenotypes in AFR and 82 in EUR, while T2D PRS correlated with 84 phenotypes in AFR and 78 in EUR. Notably, associations like hyperlipidemia, renal failure, atrial fibrillation, coronary atherosclerosis, obesity, and hypertension were observed across different PRSs in both AFR and EUR groups, with varying effect sizes and significance levels. However, in AFR individuals, the strength and number of PRS associations with other phenotypes were generally reduced compared to EUR individuals. Our study underscores the need for future research to prioritize 1) conducting GWAS in diverse ancestry groups and 2) creating a cosmopolitan PRS methodology that is universally applicable across all genetic backgrounds. Such advances will foster a more equitable and personalized approach to precision medicine.

Data from digital health technologies (DHT), including wearable sensors like Apple Watch, Whoop, Oura Ring, and Fitbit, are increasingly being used in biomedical research. Research and development of DHT-related devices, platforms, and applications is happening rapidly and with significant private-sector involvement with new biotech companies and large tech companies (e.g. Google, Apple, Amazon, Uber) investing heavily in technologies to improve human health. Many academic institutions are building capabilities related to DHT research, often in cross-sector collaboration with technology companies and other organizations with the goal of generating clinically meaningful evidence to improve patient care, to identify users at an earlier stage of disease presentation, and to support health preservation and disease prevention. Large research consortia, cross-sector partnerships, and individual research labs are all represented in the current corpus of published studies. Some of the large research studies, like NIH's All of Us Research Program, make data sets from wearable sensors available to the research community, while the vast majority of data from wearable sensors and other DHTs are held by private sector organizations and are not readily available to the research community. As data are unlocked from the private sector and made available to the academic research community, there is an opportunity to develop innovative analytics and methods through expanded access. This is the second year for this Session which solicited research results leveraging digital health technologies, including wearable sensor data, describing novel analytical methods, and issues related to diversity, equity, inclusion (DEI) of the research, data, and the community of researchers working in this area. We particularly encouraged submissions describing opportunities for expanding and democratizing academic research using data from wearable sensors and related digital health technologies.

The quickly-expanding nature of published medical literature makes it challenging for clinicians and researchers to keep up with and summarize recent, relevant findings in a timely manner. While several closed-source summarization tools based on large language models (LLMs) now exist, rigorous and systematic evaluations of their outputs are lacking. Furthermore, there is a paucity of high-quality datasets and appropriate benchmark tasks with which to evaluate these tools. We address these issues with four contributions: we release Clinfo.ai, an open-source WebApp that answers clinical questions based on dynamically retrieved scientific literature; we specify an information retrieval and abstractive summarization task to evaluate the performance of such retrieval-augmented LLM systems; we release a dataset of 200 questions and corresponding answers derived from published systematic reviews, which we name PubMed Retrieval and Synthesis (PubMedRS-200); and report benchmark results for Clinfo.ai and other publicly available OpenQA systems on PubMedRS-200.

Immune modulation is considered a hallmark of cancer initiation and progression, with immune cell density being consistently associated with clinical outcomes of individuals with cancer. Multiplex immunofluorescence (mIF) microscopy combined with automated image analysis is a novel and increasingly used technique that allows for the assessment and visualization of the tumor microenvironment (TME). Recently, application of this new technology to tissue microarrays (TMAs) or whole tissue sections from large cancer studies has been used to characterize different cell populations in the TME with enhanced reproducibility and accuracy. Generally, mIF data has been used to examine the presence and abundance of immune cells in the tumor and stroma compartments; however, this aggregate measure assumes uniform patterns of immune cells throughout the TME and overlooks spatial heterogeneity. Recently, the spatial contexture of the TME has been explored with a variety of statistical methods. In this PSB workshop, speakers will present some of the state-of-the-art statistical methods for assessing the TIME from mIF data.

Recent research has effectively used quantitative traits from imaging to boost the capabilities of genome-wide association studies (GWAS), providing further understanding of disease biology and various traits. However, it's important to note that phenotyping inherently carries measurement error and noise that could influence subsequent genetic analyses. The study focused on left ventricular ejection fraction (LVEF), a vital yet potentially inaccurate quantitative measurement, to investigate how imprecision in phenotype measurement affects genetic studies. Several methods of acquiring LVEF, along with simulating measurement noise, were assessed for their effects on ensuing genetic analyses. The results showed that by introducing just 7.9% of measurement noise, all genetic associations in an LVEF GWAS with almost forty thousand individuals could be eliminated. Moreover, a 1% increase in mean absolute error (MAE) in LVEF had an effect equivalent to a 10% reduction in the sample size of the cohort on the power of GWAS. Therefore, enhancing the accuracy of phenotyping is crucial to maximize the effectiveness of genome-wide association studies.