Biology subjects

Long, D. S.

Publications and source records attributed to Long, D. S..

4 recordsLinked to original sources

bioRxiv2022

A Machine Learning Framework to Predict Subcellular Morphology of Endothelial Cells for Digital Twin Generation

Gaining insight into different cell behaviors is key to better understanding different pathologies. These behaviors may be explained in part through close observation of 3D cell morphology. Therefore, the objective of this research was to develop a machine learning (ML) framework that can predict 3D subcellular morphological variation of endothelial cells (ECs) to generate digital twins. ECs were cultured and their membrane, nucleus, and focal adhesion (FA) sites were stained and imaged with confocal microscopy. The multicellular confocal z stacks were segmented resulting in a total of 60 single-cell stacks. Fifty randomly picked cells were augmented 20-fold to train the ML framework, and the remaining 10 were used for an independent test of prediction accuracy. The ML framework was based on an open-source conditional generative adversarial network (cGAN), which was expanded to make 3D predictions using membrane only as input to predict nucleus and FA morphology. After training the framework, the results on the independent test showed an average prediction accuracy of [~]87% for nucleus and [~]70% for FA sites. The predictions were used to build a digital twin of each EC and compared to their respective ground truth, showing an average [~]79% global accuracy and [~]84% accuracy in FA-Nucleus distribution. The results presented show the effectiveness of the developed ML framework to generate digital twins of Ecs using limited amount of data. These digital twins can be used to couple EC morphology with different behaviors. The ML framework can be potentially expanded to predict morphology of other subcellular structures as well as to study other types of cells.

bioengineering↗

bioRxiv2021

Discrete Protein Metric (DPM): A new image similarity metric to calculate accuracy of deep learning-generated cell focal adhesion predictions

Understanding cell behaviors can provide new knowledge on the development of different pathologies. Focal adhesion (FA) sites are important sub-cellular structures that are involved in these processes. To better facilitate the study of FA sites, deep learning (DL) can be used to predict FA site morphology based on limited datasets (e.g., cell membrane images). However, calculating the accuracy score of these predictions can be challenging due to the discrete/point pattern like nature of FA sites. In the present work, a new image similarity metric, discrete protein metric (DPM), was developed to calculate FA prediction accuracy. This metric measures differences in distribution (d), shape/size (s), and angle (a) of FA sites between the predicted image and its ground truth image. Performance of the DPM was evaluated by comparing it to three other commonly used image similarity metrics: Pearson correlation coefficient (PCC), feature similarity index (FSIM), and Intersection over Union (IoU). A sensitivity analysis was performed by comparing changes in each metric value due to quantifiable changes in FA site location, number, aspect ratio, area, or orientation. Furthermore, accuracy score of DL-generated predictions was calculated using all four metrics to compare their ability to capture variation across samples. Results showed better sensitivity and range of variation for DPM compared to the other metrics tested. Most importantly, DPM had the ability to determine which FA predictions were quantitatively more accurate and consistent with qualitative assessments. The proposed DPM hence provides a method to validate DL-generated FA predictions and can be extended to evaluating other predicted or segmented discrete structures of biomedical relevance.

bioengineering↗

bioRxiv2020

One step forward, two steps back: transcriptional advancements and fermentation phenomena in Actinobacillus succinogenes 130Z

Within the field of bioproduction, non-model organisms offer promise as bio-platform candidates. Non-model organisms can possess natural abilities to consume complex feedstocks, produce industrially useful chemicals, and withstand extreme environments that can be ideal for product extraction. However, non-model organisms also come with unique challenges due to lack of characterization. As a consequence, developing synthetic biology tools, predicting growth behavior, and building computational models can be difficult. There have been many advancements that have improved work with non-model organisms to address broad limitations, however each organism can come with unique surprises. Here we share our work in the non-model bacterium Actinobacillus succinognes 130Z, which includes both advancements in synthetic biology toolkit development and pitfalls in unpredictable fermentation behaviors. To develop a synthetic biology "tool kit" for A. succinogenes, information gleaned from a growth study and antibiotic screening was used to characterize 22 promoters which demonstrated a 260-fold range of fluorescence protein expression. The strongest of the promoters was incorporated into an inducible system for tunable gene control in A. succinogenes using the promoter for the lac operon as a template. This system flaunted a 481-fold range of expression and no significant basal expression. These findings were accompanied by unexpected changes in fermentation products characterized by a loss of succinic acid and increase in lactic acid after approximately 10 months in the lab. Contamination and mutation were ruled out as causes and further testing is needed to elucidate the driving factors. The significance of this work is to share tools developed in A. succinogenes while simultaneously serving as a cautionary tale. In sharing our findings, we seek to provide necessary information for further development of A. succinogenes as a platform for bioproduction of succinic acid. Additionally, we hope to illustrate the importance of diligent and long-term observation when working with non-model bacteria.

bioengineering↗

bioRxiv2018

Migration through a small pore disrupts inactive chromatin organisation in neutrophil-like cells

BackgroundMammalian cells are flexible and can rapidly change shape when they contract, adhere, or migrate. Their nucleus must be stiff enough to withstand cytoskeletal forces, but flexible enough to remodel as the cell changes shape. This is particularly important for cells migrating through constricted space, where the nuclear shape must change in order to fit through the constriction. This occurs many times in the life cycle of a neutrophil, which must protect its chromatin from damage and disruption associated with migration.\n\nResultsTotal RNA-sequencing identified that neutrophil migration through 5 or 14m pores was associated with changes in the transcript levels of inflammation and chemotaxis-related genes, when compared to unmigrated cells. Differentially expressed transcripts specific to migration with constriction were enriched for groups of genes associated with cytoskeletal remodeling.\n\nHi-C was used to capture the genome organization in control and migrated cells. Minimal switching was observed between the active (A) and inactive (B) compartments after migration. However, global depletion of short range contacts was observed following migration with constriction compared to migration without constriction. Regions with disrupted contacts, TADs, and compartments were enriched for inactive chromatin.\n\nConclusionShort range genome organization is preferentially altered in inactive chromatin, possibly protecting transcriptionally active contacts from the disruptive effects of migration with constriction. This is consistent with current hypotheses implicating heterochromatin as the mechanoresponsive form of chromatin. Further investigation concerning the contribution of heterochromatin to stiffness, flexibility, and protection of nuclear function will be important for understanding cell migration in human health and disease.

genomics↗