Search NASA⌕ Search

Engineering topics

Claborne, Daniel M.

Publications and source records attributed to Claborne, Daniel M..

OmicsMLMentor: A Web Application for Guided Machine Learning Analysis of Omics Data

Expression-based omics technologies (e.g. proteomics, metabolomics, transcriptomics, etc.) increasingly rely on supervised and unsupervised machine learning (ML) models to find key biomolecules distinguishing conditions, identify natural groupings in biological data, or generate predictions for outcomes of interest. Fitting ML models to omics data presents several challenges, including handling missing data, selecting a normalization method, choosing a valid model, and optimizing hyperparameters, all requiring statistical programming skills to address these challenges. Thus, the open-source web application SLOPE was designed to lower the barrier to ML modeling for omics data. SLOPE supports the fitting of 15 ML models (10 supervised and 5 unsupervised) tailored to omics datasets, such as proteomics, metabolomics, lipidomics, and transcriptomics. SLOPE offers several omics-specific features, including methods for handling missingness (imputation, conversion, removal), normalization tests, ranking of models based on the structure of a user’s data and user input, and optimal hyperparameter selections using cross-validation splits. By streamlining ML workflows for omics analysis, SLOPE address critical gaps in existing online web tools, facilitating a broader adoption of these models for omics research. Here, SLOPE is applied to data from a lignin exposure study to highlight the workflow for fitting both supervised and unsupervised models to data.

lipidomics↗

Missing data in multi-omics integration: Recent advances through artificial intelligence

Biological systems function through complex interactions between various ‘omics (biomolecules), and a more complete understanding of these systems is only possible through an integrated, multi-omic perspective. This has presented the need for the development of integration approaches that are able to capture the complex, often non-linear, interactions that define these biological systems and are adapted to the challenges of combining the heterogenous data across ‘omic views. A principal challenge to multi-omic integration is missing data because all biomolecules are not measured in all samples. Due to either cost, instrument sensitivity, or other experimental factors, data for a biological sample may be missing for one or more ‘omic techologies. Recent methodological developments in artificial intelligence and statistical learning have greatly facilitated the analyses of multi-omics data, however many of these techniques assume access to completely observed data. A subset of these methods incorporate mechanisms for handling partially observed samples, and these methods are the focus of this review. We describe recently developed approaches, noting their primary use cases and highlighting each method's approach to handling missing data. We additionally provide an overview of the more traditional missing data workflows and their limitations; and we discuss potential avenues for further developments as well as how the missing data issue and its current solutions may generalize beyond the multi-omics context.

97 MATHEMATICS AND COMPUTING↗

Simulated Night- Shift Schedule Disrupts the Plasma Lipidome and Reveals Early Markers of Cardiovascular Disease Risk

Introduction: The circadian system coordinates daily rhythms in lipid metabolism, storage and utilization. Disruptions of internal circadian rhythms due to altered sleep/wake schedules, such as in night-shift work, have been implicated in increased risk of cardiovascular disease and metabolic disorders. To determine the impact of a night-shift schedule on the human blood plasma lipidome, an in-laboratory simulated shift work study was conducted. Methods: Fourteen healthy young adults were assigned to 3 days of either a simulated day or night-shift schedule, followed by a 24-h constant routine protocol with fixed environmental conditions, hourly isocaloric snacks, and constant wakefulness to investigate endogenous circadian rhythms. Blood plasma samples collected at 3-h intervals were subjected to untargeted lipidomics analysis. Results: More than 400 lipids were identified and quantified across 21 subclasses. Focusing on lipids with low between-subject variation per shift condition, alterations in the circulating plasma lipidome revealed generally increased mean triglyceride levels and decreased mean phospholipid levels after night-shift relative to day-shift. The circadian rhythms of triglycerides containing odd chain fatty acids peaked earlier during constant routine after night-shift. Regardless of shift condition, triglycerides tended to either peak or be depleted at 16:30 h, with chain-specific differences associated with the direction of change. Discussion: The simulated night-shift schedule was associated with altered temporal patterns in the lipidome. This may be premorbid to the elevated cardiovascular risk that has been found epidemiologically in night-shift workers.

59 BASIC BIOLOGICAL SCIENCES↗

Video Summarization Using Deep Action Recognition Features and Robust Principal Components Analysis

In an instance where desired pre-defined actions, behaviors, or other categories are known a priori, various video classification and recognition models can be trained to discover those classifications and their location within the video. Absent that information, one might still be tasked with identifying interesting portions within a video, a process which—if done manually—is onerous and time-consuming as it requires manual inspection of the video itself. Recognizing high-level interesting segments within a whole video has been a general area of interest due to the ubiquity of video data. However the size of the data makes storage, retrieval, and inspection of large collections of videos cumbersome. This problem motivates the task of generating shortened clips highlighting the primary content of a video, relieving the burden of having to watch the entire video. This paper presents an unsupervised method of creating shortened clips of videos, enabling the rapid review of the most interesting content within a video. Our method uses features extracted from pre-trained action recognition models as input to online moving window robust principal component analysis to generate summaries. The procedure is tested on a publicly available video summarization dataset and demonstrates comparable performance to state-of-the-art in an un-augmented setting while requiring no training.

Claborne, Daniel M.↗