In line with recent advances in neural drug design and sensitivity prediction, we propose a novel architecture for interpretable prediction of anticancer compound sensitivity using a multimodal ...attention-based convolutional encoder. Our model is based on the three key pillars of drug sensitivity: compounds’ structure in the form of a SMILES sequence, gene expression profiles of tumors, and prior knowledge on intracellular interactions from protein–protein interaction networks. We demonstrate that our multiscale convolutional attention-based encoder significantly outperforms a baseline model trained on Morgan fingerprints and a selection of encoders based on SMILES, as well as the previously reported state-of-the-art for multimodal drug sensitivity prediction (R 2 = 0.86 and RMSE = 0.89). Moreover, the explainability of our approach is demonstrated by a thorough analysis of the attention weights. We show that the attended genes significantly enrich apoptotic processes and that the drug attention is strongly correlated with a standard chemical structure similarity index. Finally, we report a case study of two receptor tyrosine kinase (RTK) inhibitors acting on a leukemia cell line, showcasing the ability of the model to focus on informative genes and submolecular regions of the two compounds. The demonstrated generalizability and the interpretability of our model testify to its potential for in silico prediction of anticancer compound efficacy on unseen cancer cells, positioning it as a valid solution for the development of personalized therapies as well as for the evaluation of candidate compounds in de novo drug design.
Genome-wide identification of the mechanism of action (MoA) of small-molecule compounds characterizing their targets, effectors, and activity modulators represents a highly relevant yet elusive goal, ...with critical implications for assessment of compound efficacy and toxicity. Current approaches are labor intensive and mostly limited to elucidating high-affinity binding target proteins. We introduce a regulatory network-based approach that elucidates genome-wide MoA proteins based on the assessment of the global dysregulation of their molecular interactions following compound perturbation. Analysis of cellular perturbation profiles identified established MoA proteins for 70% of the tested compounds and elucidated novel proteins that were experimentally validated. Finally, unknown-MoA compound analysis revealed altretamine, an anticancer drug, as an inhibitor of glutathione peroxidase 4 lipid repair activity, which was experimentally confirmed, thus revealing unexpected similarity to the activity of sulfasalazine. This suggests that regulatory network analysis can provide valuable mechanistic insight into the elucidation of small-molecule MoA and compound similarity.
Display omitted
•DeMAND—a method to predict genes involved in mechanism of action of a compound•DeMAND predictions can be used to identify compound similarity•Known MoA genes are identified with high precision, sensitivity, and specificity•Novel predictions of both MoA and similarity were experimentally validated
The mechanism of action (MoA) of small-molecule compounds is elucidated by analyzing regulatory networks to identify proteins whose interactions are affected following compound perturbation. Experimental validation of novel MoA predictions revealed that the anticancer drug altretamine acts as an inhibitor of GPX4 lipid repair activity, revealing unexpected similarity to the activity of sulfasalazine.
The field of learning design studies how to support teachers in devising suitable activities for their students to learn. The field of learning analytics explores how data about students' ...interactions can be used to increase the understanding of learning experiences. Despite its clear synergy, there is only limited and fragmented work exploring the active role that data analytics can play in supporting design for learning. This paper builds on previous research to propose a framework (analytics layers for learning design) that articulates three layers of data analytics-learning analytics, design analytics and community analytics-to support informed decision-making in learning design. Additionally, a set of tools and experiences are described to illustrate how the different data analytics perspectives proposed by the framework can support learning design processes. Author abstract
Abstract
The identification of new targeted and personalized therapies for cancer requires the fast and accurate assessment of the drug efficacy of potential compounds against a particular ...biomolecular sample. It has been suggested that the integration of complementary sources of information might strengthen the accuracy of a drug efficacy prediction model. Here, we present a web-based platform for the Prediction of AntiCancer Compound sensitivity with Multimodal Attention-based Neural Networks (PaccMann). PaccMann is trained on public transcriptomic cell line profiles, compound structure information and drug sensitivity screenings, and outperforms state-of-the-art methods on anticancer drug sensitivity prediction. On the open-access web service (https://ibm.biz/paccmann-aas), users can select a known drug compound or design their own compound structure in an interactive editor, perform in-silico drug testing and investigate compound efficacy on publicly available or user-provided transcriptomic profiles. PaccMann leverages methods for model interpretability and outputs confidence scores as well as attention heatmaps that highlight the genes and chemical sub-structures that were more important to make a prediction, hence facilitating the understanding of the model’s decision making and the involved biochemical processes. We hope to serve the community with a toolbox for fast and efficient validation in drug repositioning or lead compound identification regimes.
Development of input connections in neural cultures Soriano, Jordi; Rodríguez Martínez, María; Tlusty, Tsvi ...
Proceedings of the National Academy of Sciences,
09/2008, Letnik:
105, Številka:
37
Journal Article
Recenzirano
Odprti dostop
We introduce an approach for the quantitative assessment of the connectivity in neuronal cultures, based on the statistical mechanics of percolation on a graph. This allows us to monitor the ...development of the culture and to see the emergence of connectivity in the network. The culture becomes fully connected at a time equivalent to the expected time of birth. The spontaneous bursting activity that characterizes cultures develops in parallel with the connectivity. The average number of inputs per neuron can be quantitatively determined in units of m₀, the number of activated inputs needed to excite the neuron. For m₀ similar, equals 15 we find that hippocampal neurons have on average almost equal to60-120 inputs, whereas cortical neurons have almost equal to75-150, depending on neuronal density. The ratio of excitatory to inhibitory neurons is determined by using the GABAA antagonist bicuculine. This ratio changes during development and reaches the final value at day 7-8, coinciding with the expected time of the GABA switch. For hippocampal cultures the inhibitory cells comprise almost equal to30% of the neurons in the culture whereas for cortical cultures they are almost equal to20%. Such detailed global information on the connectivity of networks in neuronal cultures is at present inaccessible by any electrophysiological or other technique.
► Gastric digestion conditions increased the release of soymilk bioactive compounds. ► Soymilk bioactive compounds content was kept or diminished though intestinal digestion. ► Despite their low ...concentration, most soymilk compounds were bioaccessible enough. ► Isoflavones were more bioaccessible than phenolic compounds.
The aim of this research was to evaluate changes in the phenolic compounds, isoflavones and antioxidant activity of soymilk following in vitro gastrointestinal digestion (including dialysis). Gastric digestion significantly influenced the release of bioactive substances from the soymilk matrix, increasing the concentration of total phenolic components (35% as the sum of individuals and 14% by Folin–Ciocalteu F–C method), total isoflavone content (22%) and total antioxidant activity (76%). The concentration of all those compounds was reduced significantly in the duodenal fraction in comparison to gastric digestion and their lowest concentration was observed in the dialysed fraction, where phenolic acids were not detected. The bioaccessibility of soymilk phenolic compounds was 15% as the sum of individuals and 20% by F–C assay; isoflavones 36% and constituents with antioxidant activity 27%. Results suggest that most of these compounds were sufficiently available to be absorbed and could contribute health benefits.
The regulatory role of nitric oxide (NO) and phytoglobins in plant response to pathogenic and mutualistic microbes has been evidenced. However, little is known about their function in the arbuscular ...mycorrhizal (AM) symbiosis. We investigated whether NO and phytoglobin PHYTOGB1 are regulatory components in the AM symbiosis.
Rhizophagus irregularis in vitro-grown cultures and tomato plants were used to monitor AM-associated NO-related root responses as compared to responses triggered by the pathogen Fusarium oxysporum. A genetic approach was conducted to understand the role of PHYTOGB1 on NO signaling during both interactions.
After a common early peak in NO levels in response to both fungi, a specific NO accumulation pattern was triggered in tomato roots during the onset of the AM interaction. PHYTOGB1 was upregulated by the AM interaction. By contrast, the pathogen triggered a continuous NO accumulation and a strong downregulation of PHYTOGB1. Manipulation of PHYTOGB1 levels in overexpressing and silenced roots led to a deregulation of NO levels and altered mycorrhization and pathogen infection.
We demonstrate that the onset of the AM symbiosis is associated with a specific NO-related signature in the host root. We propose that NO regulation by PHYTOGB1 is a regulatory component of the AM symbiosis.
The aim of this research was to evaluate the influence of an in vitro gastrointestinal digestion on the stability and bioaccessibility of vitamin C, phenolic compounds, and carotenoids, as well as ...the antioxidant activity in a blended fruit juice (BFJ) containing orange, pineapple, and kiwi. Vitamin C and most of the analyzed phenolic compounds were quite stable under gastric conditions (recovery > 75%), whereas carotenoids diminished significantly (to 64%). The concentration of all the evaluated compounds decreased during small intestinal digestion. The bioaccessibility of hydrophilic constituents was higher than that of lipophilic constituents. Flavonoids, vitamin C, and phenolic acids showed bioaccessibilities of 20.1, 15.0, and 12.7%, respectively. However, carotenes and xanthophylls were around 7.6 and 17.4% available for absorption. Despite the decrease in the concentration of these bioactive compounds after being subjected to an in vitro gastrointestinal digestion, results suggest that BFJ is an important source of bioaccessible constituents.
Advancements in mass spectrometry‐based proteomics have enabled experiments encompassing hundreds of samples. While these large sample sets deliver much‐needed statistical power, handling them ...introduces technical variability known as batch effects. Here, we present a step‐by‐step protocol for the assessment, normalization, and batch correction of proteomic data. We review established methodologies from related fields and describe solutions specific to proteomic challenges, such as ion intensity drift and missing values in quantitative feature matrices. Finally, we compile a set of techniques that enable control of batch effect adjustment quality. We provide an R package, "proBatch", containing functions required for each step of the protocol. We demonstrate the utility of this methodology on five proteomic datasets each encompassing hundreds of samples and consisting of multiple experimental designs. In conclusion, we provide guidelines and tools to make the extraction of true biological signal from large proteomic studies more robust and transparent, ultimately facilitating reliable and reproducible research in clinical proteomics and systems biology.
In mass spectrometry‐based proteomics, handling large sample sets introduces technical variability known as batch effects. This tutorial provides guidelines and tools for the assessment, normalization, and batch correction of proteomics data.
Students with special education have difficulties to develop cognitive abilities and acquire new knowledge. They could also need to improve their behavior, communication and relationships with their ...environment. The development of customizable and adaptable applications tailored to them provides many benefits as it helps mold the learning process to different cognitive, sensorial or mobility impairments. We have devised a mobile platform (based on iPad and iPod touch devices), called Picaa and designed to cover the main phases of the learning process: preparation, use and evaluation. It includes four kinds of educational activities (Exploration, Association, Puzzle and Sorting), which can be personalized by educators at content and user interface levels through a design mainly centered on student requirements, whose user profiles can also be adapted. We have performed a pre-experimental study about the use of Picaa by 39 students with special education needs from Spain, including an evaluation based on pre/post testing. The use of the learning platform Picaa is associated with positive effects in the development of learning skills for children who have special educational needs, observing that the basic skills (language, math, environmental awareness, autonomy and social) have been improved. Besides, in many cases they have the opportunity to perform activities that previously were not accessible to them, because of the interface and contents of the activities have been adapted specifically to them. The study also suggests that the repertoire of types of activities provided is suitable for learning purposes with students with impairments. Finally, the use of electronic devices and multimedia contents increases their interest in learning and attention.
► Customizable applications help mold the learning process to students' impairments. ► Mobile technologies provide autonomy and ubiquity to the learning process. ► A mobile learning platform to support students with special needs is proposed. ► An experimental study with students with special educational needs is described. ► Study results suggest positive effects on the development of learning skills.