Expanding hematopoietic stem cell ex vivo: recent advances and technical considerations.

Hematopoietic stem cells (HSCs) are the most primitive cell type in the hematopoietic hierarchy, which are responsible for sustaining the lifelong production of mature blood and immune cells. Due to their superior long-term regenerative capacity, HSC therapies such as stem cell transplantation have been used in a broad range of hematologic disorders. However, the rarity of this population in vivo considerably limits its clinical applications and large-scale analyses such as screening and safety studies. Therefore, ex vivo culture methods that allow long-term expansion and maintenance of functional HSCs are instrumental in overcoming the difficulties in studying HSC biology and improving HSC therapies. In this perspective, we discuss recent advances and technical considerations for three ex vivo HSC expansion methods including 1) polyvinyl alcohol-based HSC expansion, 2) mesenchymal stromal cell-HSC co-culture, and 3) two-/three-dimensional hydrogel HSC culture. This review summarizes the presentations and discussions from the 2022 International Society for Experimental Hematology (ISEH) Annual Meeting New Investigator Technology Session.

Towards the portability of knowledge in reinforcement learning-based systems for automatic drone navigation.

In the field of artificial intelligence (AI) one of the main challenges today is to make the knowledge acquired when performing a certain task in a given scenario applicable to similar yet different tasks to be performed with a certain degree of precision in other environments. This idea of knowledge portability is of great use in Cyber-Physical Systems (CPS) that face important challenges in terms of reliability and autonomy. This article presents a CPS where unmanned vehicles (drones) are equipped with a reinforcement learning system so they may automatically learn to perform various navigation tasks in environments with physical obstacles. The implemented system is capable of isolating the agents' knowledge and transferring it to other agents that do not have prior knowledge of their environment so they may successfully navigate environments with obstacles. A complete study has been performed to ascertain the degree to which the knowledge obtained by an agent in a scenario may be successfully transferred to other agents in order to perform tasks in other scenarios without prior knowledge of the same, obtaining positive results in terms of the success rate and learning time required to complete the task set in each case. In particular, those two indicators showed better results (higher success rate and lower learning time) with our proposal compared to the baseline in 47 out of the 60 tests conducted (78.3%).

Improving the portability of predicting students' performance models by using ontologies.

One of the main current challenges in Educational Data Mining and Learning Analytics is the portability or transferability of predictive models obtained for a particular course so that they can be applied to other different courses. To handle this challenge, one of the foremost problems is the models' excessive dependence on the low-level attributes used to train them, which reduces the models' portability. To solve this issue, the use of high-level attributes with more semantic meaning, such as ontologies, may be very useful. Along this line, we propose the utilization of an ontology that uses a taxonomy of actions that summarises students' interactions with the Moodle learning management system. We compare the results of this proposed approach against our previous results when we used low-level raw attributes obtained directly from Moodle logs. The results indicate that the use of the proposed ontology improves the portability of the models in terms of predictive accuracy. The main contribution of this paper is to show that the ontological models obtained in one source course can be applied to other different target courses with similar usage levels without losing prediction accuracy.

Data science for analyzing and improving educational processes.

In this full review paper, the recent emerging trends in Educational Data Science have been reviewed and explored to address the recent topics and contributions in the era of Smart Education. This includes a set of rigorously reviewed world-class manuscripts addressing and detailing state-of-the-art, frameworks and techniques research projects in the area of Data Science applied to Education, using different approaches such as Information Fusion, Soft Computing, Machine Learning, and Internet of Things, among others. Based on this systematic review, we have put some recommendations and suggestions for researchers, practitioners and scholars to improve their research quality in this area.

Exploiting Single-Cell Tools in Gene and Cell Therapy.

Single-cell molecular tools have been developed at an incredible pace over the last five years as sequencing costs continue to drop and numerous molecular assays have been coupled to sequencing readouts. This rapid period of technological development has facilitated the delineation of individual molecular characteristics including the genome, transcriptome, epigenome, and proteome of individual cells, leading to an unprecedented resolution of the molecular networks governing complex biological systems. The immense power of single-cell molecular screens has been particularly highlighted through work in systems where cellular heterogeneity is a key feature, such as stem cell biology, immunology, and tumor cell biology. Single-cell-omics technologies have already contributed to the identification of novel disease biomarkers, cellular subsets, therapeutic targets and diagnostics, many of which would have been undetectable by bulk sequencing approaches. More recently, efforts to integrate single-cell multi-omics with single cell functional output and/or physical location have been challenging but have led to substantial advances. Perhaps most excitingly, there are emerging opportunities to reach beyond the description of static cellular states with recent advances in modulation of cells through CRISPR technology, in particular with the development of base editors which greatly raises the prospect of cell and gene therapies. In this review, we provide a brief overview of emerging single-cell technologies and discuss current developments in integrating single-cell molecular screens and performing single-cell multi-omics for clinical applications. We also discuss how single-cell molecular assays can be usefully combined with functional data to unpick the mechanism of cellular decision-making. Finally, we reflect upon the introduction of spatial transcriptomics and proteomics, its complementary role with single-cell RNA sequencing (scRNA-seq) and potential application in cellular and gene therapy.

Ultraflat Gold QCM Electrodes Fabricated with Pressure-Forming Template Stripping for Protein Studies at the Nanoscale.

Single-molecule imaging of proteins using atomic force microscopy (AFM) is crucially dependent on protein attachment to ultraflat substrates. The template-stripping (TS) technique, which can be used to create large areas of atomically flat gold, has been used to great effect for this purpose. However, this approach requires an epoxy, which can swell in solution, causing surface roughening and substantially increasing the thickness of any sample, preventing its use on acoustic resonators in liquid. Diffusion bonding techniques should circumvent this problem but cannot be used on samples containing patterned features with mismatched heights because of cracking and poor transfer. Here, we describe a new technique called pressure-forming TS (PTS), which permits an ultraflat (0.35 ± 0.05 nm root-mean-square roughness) layer of gold to be transferred to the surface of a patterned substrate at low temperature and pressure. We demonstrate this technique by modifying a quartz crystal microbalance (QCM) sensor to contain an ultraflat gold surface. Standard QCM chips have substantial roughness, preventing AFM imaging of proteins on the surface after measurement. With our approach, there is no need to run samples in parallel: the modified QCM chip is flat enough to permit high-contrast AFM imaging after adsorption studies have been conducted. The PTS-QCM chips are then used to demonstrate adsorption of bovine serum albumin in comparison to rough QCM chips. The ability to attach thin layers of ultraflat metals to surfaces of heterogeneous nature without epoxy will have many applications in diverse fields where there is a requirement to observe nanoscale phenomena with multiple techniques, including surface and interfacial science, optics, and biosensing.

Identification and Characterization of Sterol Acyltransferases Responsible for Steryl Ester Biosynthesis in Tomato.

Steryl esters (SEs) serve as a storage pool of sterols that helps to maintain proper levels of free sterols (FSs) in cell membranes throughout plant growth and development, and participates in the recycling of FSs and fatty acids released from cell membranes in aging tissues. SEs are synthesized by sterol acyltransferases, a family of enzymes that catalyze the transfer of fatty acil groups to the hydroxyl group at C-3 position of the sterol backbone. Sterol acyltransferases are categorized into acyl-CoA:sterol acyltransferases (ASAT) and phospholipid:sterol acyltransferases (PSAT) depending on whether the fatty acyl donor substrate is a long-chain acyl-CoA or a phospolipid. Until now, only Arabidopsis ASAT and PSAT enzymes (AtASAT1 and AtPSAT1) have been cloned and characterized in plants. Here we report the identification, cloning, and functional characterization of the tomato (Solanum lycopersicum cv. Micro-Tom) orthologs. SlPSAT1 and SlASAT1 were able to restore SE to wild type levels in the Arabidopsis psat1-2 and asat1-1 knock-out mutants, respectively. Expression of SlPSAT1 in the psat1-2 background also prevented the toxicity caused by an external supply of mevalonate and the early senescence phenotype observed in detached leaves of this mutant, whereas expression of SlASAT1 in the asat1-1 mutant revealed a clear substrate preference of the tomato enzyme for the sterol precursors cycloartenol and 24-methylene cycloartanol. Subcellular localization studies using fluorescently tagged SlPSAT1 and SlASAT1 proteins revealed that SlPSAT1 localize in cytoplasmic lipid droplets (LDs) while, in contrast to the endoplasmic reticulum (ER) localization of AtASAT1, SlASAT1 resides in the plasma membrane (PM). The possibility that PM-localized SlASAT1 may act catalytically in trans on their sterol substrates, which are presumably embedded in the ER membrane, is discussed. The widespread expression of SlPSAT1 and SlASAT1 genes in different tomato organs together with their moderate transcriptional response to several stresses suggests a dual role of SlPSAT1 and SlASAT1 in tomato plant and fruit development and the adaptive responses to stress. Overall, this study contributes to enlarge the current knowledge on plant sterol acyltransferases and set the basis for further studies aimed at understanding the role of SE metabolism in tomato plant growth and development.

Tomato UDP-Glucose Sterol Glycosyltransferases: A Family of Developmental and Stress Regulated Genes that Encode Cytosolic and Membrane-Associated Forms of the Enzyme.

Sterol glycosyltransferases (SGTs) catalyze the glycosylation of the free hydroxyl group at C-3 position of sterols to produce sterol glycosides. Glycosylated sterols and free sterols are primarily located in cell membranes where in combination with other membrane-bound lipids play a key role in modulating their properties and functioning. In contrast to most plant species, those of the genus Solanum contain very high levels of glycosylated sterols, which in the case of tomato may account for more than 85% of the total sterol content. In this study, we report the identification and functional characterization of the four members of the tomato (Solanum lycopersicum cv. Micro-Tom) SGT gene family. Expression of recombinant SlSGT proteins in E. coli cells and N. benthamiana leaves demonstrated the ability of the four enzymes to glycosylate different sterol species including cholesterol, brassicasterol, campesterol, stigmasterol, and ß-sitosterol, which is consistent with the occurrence in their primary structure of the putative steroid-binding domain found in steroid UDP-glucuronosyltransferases and the UDP-sugar binding domain characteristic for a superfamily of nucleoside diphosphosugar glycosyltransferases. Subcellular localization studies based on fluorescence recovery after photobleaching and cell fractionation analyses revealed that the four tomato SGTs, like the Arabidopsis SGTs UGT80A2 and UGT80B1, localize into the cytosol and the PM, although there are clear differences in their relative distribution between these two cell fractions. The SlSGT genes have specialized but still largely overlapping expression patterns in different organs of tomato plants and throughout the different stages of fruit development and ripening. Moreover, they are differentially regulated in response to biotic and abiotic stress conditions. SlSGT4 expression increases markedly in response to osmotic, salt, and cold stress, as well as upon treatment with abscisic acid and methyl jasmonate. Stress-induced SlSGT2 expression largely parallels that of SlSGT4. On the contrary, SlSGT1 and SlSGT3 expression remains almost unaltered under the tested stress conditions. Overall, this study contributes to broaden the current knowledge on plant SGTs and provides support to the view that tomato SGTs play overlapping but not completely redundant biological functions involved in mediating developmental and stress responses.

A Reinforcement Learning Model Equipped with Sensors for Generating Perception Patterns: Implementation of a Simulated Air Navigation System Using ADS-B (Automatic Dependent Surveillance-Broadcast) Technology.

Over the last few decades, a number of reinforcement learning techniques have emerged, and different reinforcement learning-based applications have proliferated. However, such techniques tend to specialize in a particular field. This is an obstacle to their generalization and extrapolation to other areas. Besides, neither the reward-punishment (r-p) learning process nor the convergence of results is fast and efficient enough. To address these obstacles, this research proposes a general reinforcement learning model. This model is independent of input and output types and based on general bioinspired principles that help to speed up the learning process. The model is composed of a perception module based on sensors whose specific perceptions are mapped as perception patterns. In this manner, similar perceptions (even if perceived at different positions in the environment) are accounted for by the same perception pattern. Additionally, the model includes a procedure that statistically associates perception-action pattern pairs depending on the positive or negative results output by executing the respective action in response to a particular perception during the learning process. To do this, the model is fitted with a mechanism that reacts positively or negatively to particular sensory stimuli in order to rate results. The model is supplemented by an action module that can be configured depending on the maneuverability of each specific agent. The model has been applied in the air navigation domain, a field with strong safety restrictions, which led us to implement a simulated system equipped with the proposed model. Accordingly, the perception sensors were based on Automatic Dependent Surveillance-Broadcast (ADS-B) technology, which is described in this paper. The results were quite satisfactory, and it outperformed traditional methods existing in the literature with respect to learning reliability and efficiency.

A general framework for time series data mining based on event analysis: application to the medical domains of electroencephalography and stabilometry.

There are now domains where information is recorded over a period of time, leading to sequences of data known as time series. In many domains, like medicine, time series analysis requires to focus on certain regions of interest, known as events, rather than analyzing the whole time series. In this paper, we propose a framework for knowledge discovery in both one-dimensional and multidimensional time series containing events. We show how our approach can be used to classify medical time series by means of a process that identifies events in time series, generates time series reference models of representative events and compares two time series by analyzing the events they have in common. We have applied our framework on time series generated in the areas of electroencephalography (EEG) and stabilometry. Framework performance was evaluated in terms of classification accuracy, and the results confirmed that the proposed schema has potential for classifying EEG and stabilometric signals. The proposed framework is useful for discovering knowledge from medical time series containing events, such as stabilometric and electroencephalographic time series. These results would be equally applicable to other medical domains generating iconographic time series, such as, for example, electrocardiography (ECG).