Conference Title: 2021 ACM/IEEE Joint Conference on Digital Libraries (JCDL) Conference Start Date: 2021, Sept. 27 Conference End Date: 2021, Sept. 30 Conference Location: Champaign, IL, USAMetadata are key descriptors of research data, particularly for researchers seeking to apply machine learning (ML) to the vast collections of digitized specimens. Unfortunately, the available metadata is often sparse and, at times, erroneous. Additionally, it is prohibitively expensive to address these limitations through traditional, manual means. This paper reports on research that applies machine-driven approaches to analyzing digitized fish images and extracting various important features from them. The digitized fish specimens are being analyzed as part of the Biology Guided Neural Networks (BGNN) initiative, which is developing a novel class of artificial neural networks using phylogenies and anatomy ontologies. Automatically generated metadata is crucial for identifying the high-quality images needed for the neural network's predictive analytics. Methods that combine ML and image informatics techniques allow us to rapidly enrich the existing metadata associated with the 7,244 images from the Illinois Natural History Survey (INHS) used in our study. Results show we can accurately generate many key metadata properties relevant to the BGNN project, as well as general image quality metrics (e.g. brightness and contrast). Results also show that we can accurately generate bounding boxes and segmentation masks for fish, which are needed for subsequent machine learning analyses. The automatic process outperforms humans in terms of time and accuracy, and provides a novel solution for leveraging digitized specimens in ML. This research demonstrates the ability of computational methods to enhance the digital library services associated with the tens of thousands of digitized specimens stored in open-access repositories worldwide.
more »
« less
Understanding the value of biobank attributes to researchers using a conjoint experiment
Biobanks are important in biomedical and public health research, and future healthcare research relies on their strength and capacity. However, there are financial challenges related to the operation of commercial biobanks and concerns around the commercialization of biobanks. Non-commercial biobanks depend on grant funding to operate and could be valuable to researchers if they can enable access to quality specimens at lower costs. The objective of this study is to estimate the value of specific biobank attributes. We used a rating-based conjoint experiment approach to study how researchers valued handling fee, access, quality, characterization, breadth of consent, access to key endemics, and time taken to fulfil requests. We found that researchers placed the greatest relative importance on the quality of specimens (26%), followed by the characterization of specimens (21%). Researchers with prior experience purchasing biological samples also valued access to key endemic in-country sites (11.6%) and low handling fees (5.5%) in biobanks.
more » « less- NSF-PAR ID:
- 10481083
- Publisher / Repository:
- Nature Publishing Group
- Date Published:
- Journal Name:
- Scientific Reports
- Volume:
- 13
- Issue:
- 1
- ISSN:
- 2045-2322
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Adam, N. ; Neuhold, E. ; Furuta, R. (Ed.)Metadata is a key data source for researchers seeking to apply machine learning (ML) to the vast collections of digitized biological specimens that can be found online. Unfortunately, the associated metadata is often sparse and, at times, erroneous. This paper extends previous research conducted with the Illinois Natural History Survey (INHS) collection (7244 specimen images) that uses computational approaches to analyze image quality, and then automatically generates 22 metadata properties representing the image quality and morphological features of the specimens. In the research reported here, we demonstrate the extension of our initial work to University of the Wisconsin Zoological Museum (UWZM) collection (4155 specimen images). Further, we enhance our computational methods in four ways: (1) augmenting the training set, (2) applying contrast enhancement, (3) upscaling small objects, and (4) refining our processing logic. Together these new methods improved our overall error rates from 4.6 to 1.1%. These enhancements also allowed us to compute an additional set of 17 image-based metadata properties. The new metadata properties provide supplemental features and information that may also be used to analyze and classify the fish specimens. Examples of these new features include convex area, eccentricity, perimeter, skew, etc. The newly refined process further outperforms humans in terms of time and labor cost, as well as accuracy, providing a novel solution for leveraging digitized specimens with ML. This research demonstrates the ability of computational methods to enhance the digital library services associated with the tens of thousands of digitized specimens stored in open-access repositories world-wide by generating accurate and valuable metadata for those repositories.more » « less
-
more » « less
Biobanks linked to electronic health records provide rich resources for health‐related research. With improvements in administrative and informatics infrastructure, the availability and utility of data from biobanks have dramatically increased. In this paper, we first aim to characterize the current landscape of available biobanks and to describe specific biobanks, including their place of origin, size, and data types. The development and accessibility of large‐scale biorepositories provide the opportunity to accelerate agnostic searches, expedite discoveries, and conduct hypothesis‐generating studies of disease‐treatment, disease‐exposure, and disease‐gene associations. Rather than designing and implementing a single study focused on a few targeted hypotheses, researchers can potentially use biobanks' existing resources to answer an expanded selection of exploratory questions as quickly as they can analyze them. However, there are many obvious and subtle challenges with the design and analysis of biobank‐based studies. Our second aim is to discuss statistical issues related to biobank research such as study design, sampling strategy, phenotype identification, and missing data. We focus our discussion on biobanks that are linked to electronic health records. Some of the analytic issues are illustrated using data from the Michigan Genomics Initiative and UK Biobank, two biobanks with two different recruitment mechanisms. We summarize the current body of literature for addressing these challenges and discuss some standing open problems. This work complements and extends recent reviews about biobank‐based research and serves as a resource catalog with analytical and practical guidance for statisticians, epidemiologists, and other medical researchers pursuing research using biobanks.
-
Abstract There is a need for new in vitro systems that enable pharmaceutical companies to collect more physiologically-relevant information on drug response in a low-cost and high-throughput manner. For this purpose, three-dimensional (3D) spheroidal models have been established as more effective than two-dimensional models. Current commercial techniques, however, rely heavily on self-aggregation of dissociated cells and are unable to replicate key features of the native tumor microenvironment, particularly due to a lack of control over extracellular matrix components and heterogeneity in shape, size, and aggregate forming tendencies. In this study, we overcome these challenges by coupling tissue engineering toolsets with microfluidics technologies to create engineered cancer microspheres. Specifically, we employ biosynthetic hydrogels composed of conjugated poly(ethylene glycol) (PEG) and fibrinogen protein (PEG-Fb) to create engineered breast and colorectal cancer tissue microspheres for 3D culture, tumorigenic characterization, and examination of potential for high-throughput screening (HTS). MCF7 and MDA-MB-231 cell lines were used to create breast cancer microspheres and the HT29 cell line and cells from a stage II patient-derived xenograft (PDX) were encapsulated to produce colorectal cancer (CRC) microspheres. Using our previously developed microfluidic system, highly uniform cancer microspheres (intra-batch coefficient of variation (CV) ≤ 5%, inter-batch CV < 2%) with high cell densities (>20×106 cells/ml) were produced rapidly, which is critical for use in drug testing. Encapsulated cells maintained high viability and displayed cell type-specific differences in morphology, proliferation, metabolic activity, ultrastructure, and overall microsphere size distribution and bulk stiffness. For PDX CRC microspheres, the percentage of human (70%) and CRC (30%) cells was maintained over time and similar to the original PDX tumor, and the mechanical stiffness also exhibited a similar order of magnitude (103 Pa) to the original tumor. The cancer microsphere system was shown to be compatible with an automated liquid handling system for administration of drug compounds; MDA-MB-231 microspheres were distributed in 384 well plates and treated with staurosporine (1 μM) and doxorubicin (10 μM). Expected responses were quantified using CellTiter-Glo® 3D, demonstrating initial applicability to HTS drug discovery. PDX CRC microspheres were treated with Fluorouracil (5FU) (10 to 500 μM) and displayed a decreasing trend in metabolic activity with increasing drug concentration. Providing a more physiologically relevant tumor microenvironment in a high-throughput and low-cost manner, the PF hydrogel-based cancer microspheres could potentially improve the translational success of drug candidates by providing more accurate in vitro prediction of in vivo drug efficacy. Citation Format: Elizabeth A. Lipke, Wen J. Seeto, Yuan Tian, Mohammadjafar Hashemi, Iman Hassani, Benjamin Anbiah, Nicole L. Habbit, Michael W. Greene, Dmitriy Minond, Shantanu Pradhan. Production of cancer tissue-engineered microspheres for high-throughput screening [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 175.more » « less
-
more » « less
Abstract Characterization of species diversity of zooplankton is key to understanding, assessing, and predicting the function and future of pelagic ecosystems throughout the global ocean. The marine zooplankton assemblage, including only metazoans, is highly diverse and taxonomically complex, with an estimated ~28,000 species of 41 major taxonomic groups. This review provides a comprehensive summary of DNA sequences for the barcode region of mitochondrial cytochrome oxidase I (COI) for identified specimens. The foundation of this summary is the MetaZooGene Barcode Atlas and Database (MZGdb), a new open-access data and metadata portal that is linked to NCBI GenBank and BOLD data repositories. The MZGdb provides enhanced quality control and tools for assembling COI reference sequence databases that are specific to selected taxonomic groups and/or ocean regions, with associated metadata (e.g., collection georeferencing, verification of species identification, molecular protocols), and tools for statistical analysis, mapping, and visualization. To date, over 150,000 COI sequences for ~ 5600 described species of marine metazoan plankton (including holo- and meroplankton) are available via the MZGdb portal. This review uses the MZGdb as a resource for summaries of COI barcode data and metadata for important taxonomic groups of marine zooplankton and selected regions, including the North Atlantic, Arctic, North Pacific, and Southern Oceans. The MZGdb is designed to provide a foundation for analysis of species diversity of marine zooplankton based on DNA barcoding and metabarcoding for assessment of marine ecosystems and rapid detection of the impacts of climate change.
