Topic Review
A Taxonomic Survey of Physics-Informed Machine Learning
Physics-informed machine learning (PIML) refers to the emerging area of extracting physically relevant solutions to complex multiscale modeling problems lacking sufficient quantity and veracity of data with learning models informed by physically relevant prior information.
  • 238
  • 20 Jun 2023
Topic Review
Alignment-Free Study of Viral Diversity
Viral sequence variation can expand the host repertoire, enhance the infection ability, and/or prevent the build-up of a long-term specific immunity by the host. The study of viral diversity is, thus, critical to understand sequence change and its implications for intervention strategies.
  • 574
  • 22 Sep 2021
Topic Review
Application of GANs in Gene Expression Data Augmentation
A generative adversarial network (GAN) is essentially a two-player game composed of a generator and a discriminator. The generator’s role is to create synthetic data, while the discriminator’s task is to distinguish between real and generated data. During the training process, the generator strives to produce data that the discriminator cannot differentiate from the real data, whereas the discriminator continually improves its ability to distinguish real from generated data. This adversarial training regimen imbues GANs with the capability to model complex data distributions and produce high-quality synthetic data. Notably, their application to gene expression data systems is a fascinating and rapidly growing focus area.
  • 174
  • 21 Jul 2023
Topic Review
Approaches to Cardiovascular and Respiratory Systems Modelling
'Medicine in silico' has been strongly encouraged due to ethical and legal limitations related to animal experiments and investigations conducted on patients. Computer models, particularly the very complex ones (virtual patients—VP), can be used in medical education and biomedical research as well as in clinical applications. Simpler patient-specific models may aid medical procedures. However, computer models are unfit for medical devices testing. Hybrid (i.e., numerical–physical) models do not have this disadvantage.
  • 257
  • 20 Jun 2022
Topic Review
Catalytic Factors Associated with Post-Traumatic Stress Disorder
Post-traumatic stress disorder (PTSD) is a complex psychological disorder that develops following exposure to traumatic events. PTSD is influenced by catalytic factors such as dysregulated hypothalamic-pituitary-adrenal (HPA) axis, neurotransmitter imbalances, and oxidative stress. Genetic variations may act as important catalysts, impacting neurochemical signaling, synaptic plasticity, and stress response systems. Understanding the intricate gene networks and their interactions is vital for comprehending the underlying mechanisms of PTSD. 
  • 123
  • 03 Aug 2023
Topic Review
Cell-Type Annotation
Multicellular organisms consist of cells that can be categorized by their function and morphology. Single-cell transcriptomics makes it possible to individually profile thousands of cells in multiple tissues and organisms within a single experiment. Determining and labeling cell types or states in single cell transcriptomic data is known as cell-type annotation or identification. Several methods are employed for cell-type annotation, including signature scoring, supervised learning, cell-integration-based label transfer, and semi-supervised annotation. Considering the lineage relationships among cell types, hierarchical classification methods are crucial for accurately identifying cell types and subtypes at an optimal clustering resolution. The use of well-curated reference datasets, implementation of quality control measures, and careful consideration of cluster resolutions heavily influence the reliability of cell-type annotation. The aim of cell-type annotation is to gain insights into cell heterogeneity in various biological processes and diseases, with the potential to drive improvements in therapeutic interventions.
  • 326
  • 08 Aug 2023
Topic Review
Classification Algorithms for Unifloral Honeys
Unifloral honeys are highly demanded by honey consumers, especially in Europe. To ensure that a honey belongs to a very appreciated botanical class, the classical methodology is palynological analysis to identify and count pollen grains. Highly trained personnel are needed to perform this task, which complicates the characterization of honey botanical origins. Organoleptic assessment of honey by expert personnel helps to confirm such classification. In this study, the ability of different machine learning (ML) algorithms to correctly classify seven types of Spanish honeys of single botanical origins (rosemary, citrus, lavender, sunflower, eucalyptus, heather and forest honeydew) was investigated comparatively. The botanical origin of the samples was ascertained by pollen analysis complemented with organoleptic assessment. Physicochemical parameters such as electrical conductivity, pH, water content, carbohydrates and color of unifloral honeys were used to build the dataset. The following ML algorithms were tested: penalized discriminant analysis (PDA), shrinkage discriminant analysis (SDA), high-dimensional discriminant analysis (HDDA), nearest shrunken centroids (PAM), partial least squares (PLS), C5.0 tree, extremely randomized trees (ET), weighted k-nearest neighbors (KKNN), artificial neural networks (ANN), random forest (RF), support vector machine (SVM) with linear and radial kernels and extreme gradient boosting trees (XGBoost). The ML models were optimized by repeated 10-fold cross-validation primarily on the basis of log loss or accuracy metrics, and their performance was compared on a test set in order to select the best predicting model. Built models using PDA produced the best results in terms of overall accuracy on the test set. ANN, ET, RF and XGBoost models also provided good results, while SVM proved to be the worst. 
  • 432
  • 05 Jul 2021
Topic Review
Commercial Targeted Libraries in Drug Design
After the identification of a biological target (enzyme, receptor, protein and so on), the focus of the early phase of drug discovery rests on the identification of leads or compounds that exhibit pharmacological activity against this specific target. Compounds of interest are most often discovered in pre-existing libraries of compounds that can be either virtual or physical. Computer-aided methods which have become increasingly important over the years in drug development utilize virtual compound libraries. While physical compound libraries reach the number of millions of molecules, virtual compound libraries created by large pharmaceutical companies can range from 107 to 1018 molecules. Investigations of these libraries identifies specific molecules, synthetic pathways and focus on a specific chemical space. Targeted libraries are often smaller and are focused towards a specific chemical space. They are created by using relevant biological information with the aim to decrease the processing time associated with larger libraries while maintaining the most relevant chemical space where lead compounds can be found. Due to the fact that they required less computational or wet-lab labor to process they have become very popular with smaller laboratories which try to compete in the drug-development sector. Many modern vendors of compounds today offer such libraries, but the quality of the procedure used to define desired chemical space and select compounds is questionable.
  • 963
  • 25 May 2022
Topic Review
Computational Biology in Drug Design
Traditional drug design requires a great amount of research time and developmental expense. Booming computational approaches, including computational biology, computer-aided drug design, and artificial intelligence, have the potential to expedite the efficiency of drug discovery by minimizing the time and financial cost. In recent years, computational approaches are being widely used to improve the efficacy and effectiveness of drug discovery and pipeline, leading to the approval of plenty of new drugs for marketing. The present review emphasizes on the applications of these indispensable computational approaches in aiding target identification, lead discovery, and lead optimization. Some challenges of using these approaches for drug design are also discussed. Moreover, researchers propose a methodology for integrating various computational techniques into new drug discovery and design.
  • 357
  • 28 Nov 2022
Topic Review
Computer-Aided Drug Discovery for SMA
Spinal muscular atrophy (SMA), one of the leading inherited causes of child mortality, is a rare neuromuscular disease arising from loss-of-function mutations of the survival motor neuron 1 (SMN1) gene, which encodes the SMN protein. When lacking the SMN protein in neurons, patients suffer from muscle weakness and atrophy, and in the severe cases, respiratory failure and death. Several therapeutic approaches show promise with human testing and three medications have been approved by the U.S. Food and Drug Administration (FDA) to date. Despite the shown promise of these approved therapies, there are some crucial limitations, one of the most important being the cost. The FDA-approved drugs are high-priced and are shortlisted among the most expensive treatments in the world. The price is still far beyond affordable and may serve as a burden for patients. The blooming of the biomedical data and advancement of computational approaches have opened new possibilities for SMA therapeutic development. 
  • 947
  • 09 Oct 2021
  • Page
  • of
  • 5