Biology subjects

Miller, J. M.

Publications and source records attributed to Miller, J. M..

28 records · Page 2Linked to original sources

bioRxiv2022

Enriching Representation Learning Using 53 Million Patient Notes through Human Phenotype Ontology Embedding

The Human Phenotype Ontology (HPO) is a dictionary of more than 15,000 clinical phenotypic terms with defined semantic relationships, developed to standardize their representation for phenotypic analysis. Over the last decade, the HPO has been used to accelerate the implementation of precision medicine into clinical practice. In addition, recent research in representation learning, specifically in graph embedding, has led to notable progress in automated prediction via learned features. Here, we present a novel approach to phenotype representation by incorporating phenotypic frequencies based on 53 million full-text health care notes from more than 1.5 million individuals. We demonstrate the efficacy of our proposed phenotype embedding technique by comparing our work to existing phenotypic similarity-measuring methods. Using phenotype frequencies in our embedding technique, we are able to identify phenotypic similarities that surpass the current computational models. In addition, we show that our embedding technique aligns with domain experts judgment at a level that exceeds their agreement. We show that our proposed technique efficiently represents complex and multidimensional phenotypes in HPO format, which can then be used as input for various downstream tasks that require deep phenotyping, including patient similarity analyses and disease trajectory prediction.

bioinformatics↗

bioRxiv2021

Identification of Low Population States in Cryo-EM Using Deep Learning

Cryo-EM has made extraordinary headway towards becoming a semi-automated, high-throughput structure determination technique. In the general workflow, high-to-medium population states are grouped into two- and three-dimensional classes, from which structures can be obtained with near-atomic resolution and subsequently analyzed to interpret function. However, low population states, which are also functionally important, are often discarded. Here, we describe a technique whereby low population states can be efficiently identified with minimal human effort via a deep convolutional neural network classifier. We use this deep learning classifier to describe a transient, low population state of bacteriophage A511 in the midst of infecting its bacterial host. This method can be used to further automate data collection and identify other functionally important low population states.

biophysics↗

bioRxiv2021

Chemically Induced Senescence in Human Stem Cell-Derived Neurons Promotes Phenotypic Presentation of Neurodegeneration

Modeling age-related neurodegenerative disorders with human stem cells is difficult due to the embryonic nature of stem cell derived neurons. We developed a chemical cocktail to induce senescence of iPSC-derived neurons to address this challenge. We first screened small molecules that induce embryonic fibroblasts to exhibit features characteristic of aged fibroblasts. We then optimized a cocktail of small molecules that induced senescence in fibroblasts and cortical neurons without causing DNA damage. The utility of the "senescence cocktail" was validated in motor neurons derived from ALS patient iPSCs which exhibited protein aggregation and axonal degeneration substantially earlier than those without cocktail treatment. Our "senescence cocktail" will likely enhance the manifestation of disease-related phenotypes in neurons derived from iPSCs, enabling the generation of reliable drug discovery platforms.

neuroscience↗

bioRxiv2020

Sleep is required for odor exposure to consolidate memory and remodel olfactory synapses.

Animals with complex nervous systems demand sleep for memory consolidation and synaptic remodeling. Here we show that though the Caenorhabditis elegans nervous system has a limited number of neurons, sleep is necessary for both processes. In addition, it is unclear in any system if sleep collaborates with experience to alter synapses between specific neurons and whether this ultimately affects behavior. C. elegans neurons have defined connections and well-described contributions to behavior. We show that spaced odor-training and post-training sleep induce long-term memory. Memory consolidation, but not acquisition, requires a pair of interneurons, the AIYs, which play a role in odor-seeking behavior. In worms that consolidate memory, both sleep and odor conditioning are required to diminish inhibitory synaptic connections between the AWC chemosensory neurons and the AIYs. Thus, we demonstrate in a living organism that sleep is required for events immediately after training that drive memory consolidation and alter synaptic structures.

neuroscience↗

bioRxiv2020

Heart Slice Culture System Reliably Demonstrates Clinical Drug-Related Cardiotoxicity

The limited availability of human heart tissue and its complex cell composition are major limiting factors for reliable testing drug efficacy, toxicity and understanding mechanism. Recently, we developed a functional human and pig heart slice biomimetic culture system that fully preserves the viability and functionality of 300 {micro}m heart slices for 6 days. Here, we tested the reliability of this culture system in delineating the mechanisms of known anti-cancer drugs that cause cardiomyopathy. We tested three anti-cancer drugs (doxorubicin, trastuzumab, and sunitinib) associated with different mechanisms leading to cardiotoxicity at three concentrations and assessed the effect of these drugs on heart slice viability, structure, function and transcriptome. Slices incubated with any of these drugs for 48 h showed significant loss in viability, cardiomyocyte structure and functionality. Mechanistically, RNA sequencing demonstrated a significant downregulation of cardiac genes and upregulation of oxidative response in doxorubicin-treated tissues. Trastuzumab treatment caused major downregulation in cardiac muscle contraction-related genes, consistent with its clinically known direct effect on cardiomyocytes. Interestingly, sunitinib treatment resulted in significant downregulation of angiogenesis-related genes in line with its mechanism of action. Heart slices are not only able to demonstrate the expected toxicity of doxorubicin and trastuzumab similar to hiPS-derived-cardiomyocytes; they are superior in detecting sunitinib cardiotoxicity phenotypes and mechanism in the clinically relevant concentration range, 100 nM - 1 {micro}M. These results indicate that heart slice tissue culture models have the potential to become a reliable platform for testing drug toxicity and mechanism of action.

cell biology↗

bioRxiv2019

Effective strategies for preventing reestablishment of malaria in areas with recent elimination and high transmission potential

Maintaining zero transmission after malaria elimination will be a challenging task for many countries where malaria is still endemic. When local transmission potential is high, and importation of malaria infections continues from neighboring areas with ongoing transmission, malaria programs must develop robust surveillance and outbreak response systems. However, the requirements for such systems remain unclear. Using an agent-based, spatial microsimulation model of two areas in southern Zambia, where elimination efforts are currently underway, we compare the ability of various routine and reactive intervention packages to maintain near-zero prevalence in the face of continued importation. We find that in formerly moderate-transmission areas, high treatment rate of symptomatic malaria is sufficient to prevent reestablishment of malaria. Routine redistributions of insecticide-treated nets and reactive case detection with antimalarial drugs cannot completely compensate for inadequate case management. In formerly high-transmission areas, excellent case management and maintenance of good bednet coverage are both required to prevent resurgence, and outbreak response with antimalarial drugs or additional vector control is also necessary. These results begin to describe the essential criteria for operations that successfully prevent reestablishment of malaria post-elimination and highlight the need for both long-term, sustainable excellence in primary care and comprehensive surveillance that feeds into rapid and flexible outbreak response.\n\nAuthor SummaryThe global community is working toward malaria elimination, but some areas will eliminate before others. Eliminated areas will need to develop intervention programs capable of preventing imported infections from leading to reestablishment, a particular challenge when transmission was previously very high. Past experience has shown that stopping elimination interventions leads to massive resurgence, but it is unclear which interventions must be continued, which can be stopped to conserve resources, and what new interventions should be deployed. Using a simulation model built to capture malaria transmission and intervention history of two areas that recently made enormous progress toward elimination, we tested how well different intervention programs were able to prevent reestablishment of malaria. We found that treating as many symptomatic cases as possible was the single most important intervention to implement. In some contexts, this intervention alone was sufficient to prevent reestablishment. Other areas with historically higher transmission required maintaining vector control to contain mosquito populations. Localized outbreak response with antimalarial drugs or additional vector control was also necessary and predicted to be a highly efficient use of resources. These findings provide quantitative guidance for policy-makers considering how to stratify eliminated areas and plan new operational modes for the post-elimination era.

ecology↗

bioRxiv2018

PABA Release from Chitosan-PCL with Induced Electric Current

Controlled drug delivery systems such as the stimulation-based biomaterial scaffolds for sequestration and release of drugs offer safety and regulated therapeutic approach. In this study, the drug: para-aminobenzoic acid (PABA) was absorbed into a crosslinked chitosan and poly(caprolactone) (PCL) hydrogel and its release kinetics quantified under different conditions. It was experimentally-observed that the higher the pH (or the more basic the pH), the slower the PABA saturation release trended over time. At the acidic environment of pH 4, PABA was released the fastest, and enhanced by the degradation of chitosan-PCL gel. When a constant electric current of 0.6 mA sa applied, PABA release was induced at pH 10. However, at pH 7, PABA was stably-bound to the chitosan-PCL matrix, with or without the external current. The selective sequestration of PABA at basic pH and its stimulated release via electric current application can be further explored for clinical translatability.

bioengineering↗

bioRxiv2018

Light Pollution Increases West Nile Virus Competence in a Ubiquitous Passerine Reservoir Species

One sentence summaryLight pollution increases host infectiousness.\n\nAbstractLight pollution is a growing problem, but its impacts on infectious disease risk have not been considered. Previous research has revealed that dim light at night (dLAN) dysregulates various immune functions and biorhythms, which hints that dLAN could change the risk of disease epidemics. Here, we demonstrate that dLAN enhances infectiousness of the house sparrow (Passer domesticus), an urban-dwelling avian host of West Nile virus (WNV). Sparrows exposed to dLAN maintained viral titers above the transmission threshold to a biting vector (105 plaque-forming units) for two days longer than controls but did not die at higher rates. A mathematical model revealed that such effects could increase WNV outbreak potential by ~41%. dLAN likely affects other host and vector traits relevant to transmission, so additional research is needed to advise management of zoonotic diseases in light polluted areas.

pathology↗

bioRxiv2017

Equivalent high-resolution identification of neuronal cell types with single-nucleus and single-cell RNA-sequencing

Transcriptional profiling of complex tissues by RNA-sequencing of single nuclei presents some advantages over whole cell analysis. It enables unbiased cellular coverage, lack of cell isolation-based transcriptional effects, and application to archived frozen specimens. Using a well-matched pair of single-nucleus RNA-seq (snRNA-seq) and single-cell RNA-seq (scRNA-seq) SMART-Seq v4 datasets from mouse visual cortex, we demonstrate that similarly high-resolution clustering of closely related neuronal types can be achieved with both methods if intronic sequences are included in nuclear RNA-seq analysis. More transcripts are detected in individual whole cells (~11,000 genes) than nuclei (~7,000 genes), but the majority of genes have similar detection across cells and nuclei. We estimate that the nuclear proportion of total cellular mRNA varies from 20% to over 50% for large and small pyramidal neurons, respectively. Together, these results illustrate the high information content of nuclear RNA for characterization of cellular diversity in brain tissues.

neuroscience↗

bioRxiv2017

Theory, practice, and conservation in the age of genomics: the Galapagos giant tortoise as a case study

High-throughput DNA sequencing allows efficient discovery of thousands of single nucleotide polymorphisms (SNPs) in non-model species. Population genetic theory predicts that this large number of independent markers should provide detailed insights into population structure, even when only a few individuals are sampled. Still, sampling design can have a strong impact on such inferences. Here, we use simulations and empirical SNP data to investigate the impacts of sampling design on estimating genetic differentiation among populations that represent three species of Galapagos giant tortoises (Chelonoidis spp.). Though microsatellite and mitochondrial DNA analyses have supported the distinctiveness of these species, a recent study called into question how well these markers matched with data from genomic SNPs, thereby questioning decades of studies in non-model organisms. Using >20,000 genome-wide SNPs from 30 individuals from three Galapagos giant tortoise species, we find distinct structure that matches the relationships described by the traditional genetic markers. Furthermore, we confirm that accurate estimates of genetic differentiation in highly structured natural populations can be obtained using thousands of SNPs and 2-5 individuals, or hundreds of SNPs and 10 individuals, but only if the units of analysis are delineated in a way that is consistent with evolutionary history. We show that the lack of structure in the recent SNP-based study was likely due to unnatural grouping of individuals and erroneous genotype filtering. Our study demonstrates that genomic data enable patterns of genetic differentiation among populations to be elucidated even with few samples per population, and underscores the importance of sampling design. These results have specific implications for studies of population structure in endangered species and subsequent management decisions.\n\n\"Modern molecular techniques provide unprecedented power to understand genetic variation in natural populations. Nevertheless, application of this information requires sound understanding of population genetics theory.\"\n\n- Fred Allendorf (2017, p. 420)

evolutionary biology↗