Drug discovery and repositioning are important processes for the pharmaceutical industry. These processes demand a high investment in resources and are time-consuming. Several strategies have been used to address this problem, including computer-aided drug design (CADD). Among CADD approaches, it is essential to highlight virtual screening (VS), an in silico approach based on computer simulation that can select organic molecules toward the therapeutic targets of interest. The techniques applied by VS are based on the structure of ligands (LBVS), receptors (SBVS), or fragments (FBVS). Regardless of the type of VS to be applied, they can be divided into categories depending on the used algorithms: similarity-based, quantitative, machine learning, meta-heuristics, and other algorithms. Each category has its objectives, advantages, and disadvantages. This review presents an overview of the algorithms used in VS, describing them and showing their use in drug design and their contribution to the drug development process.
Alzheimer’s Disease (AD) is a neurodegenerative condition characterized by progressive memory loss and other affected cognitive functions. Pharmacological therapy of AD relies on inhibitors of the enzymes acetylcholinesterase (AChE) and butyrylcholinesterase (BChE), offering only a palliative effect and being incapable of stopping or reversing the neurodegenerative process. However, recent studies have shown that inhibiting the enzyme -secretase 1 (BACE-1) may be able to stop neurodegeneration, making it a promising target. Considering these three enzymatic targets, it becomes feasible to apply computational techniques to guide the identification and planning of molecules capable of binding to all of them. After virtually screening 2119 molecules from a library, 13 hybrids were built and further screened by triple pharmacophoric model, molecular docking, and molecular dynamics (t = 200 ns). The selected hybrid G meets all stereo-electronic requirements to bind to AChE, BChE, and BACE-1 and offers a promising structure for future synthesis, enzymatic testing, and validation.
Alzheimer’s disease requires therapeutic strategies targeting multiple pathological mechanisms. This study investigates the neuroprotective potential of quercetin using an integrated approach combining in silico modeling and in vivo validation. Computational analyses examined the binding behavior of quercetin toward three key enzymes implicated in disease pathophysiology: acetylcholinesterase, butyrylcholinesterase, and beta-secretase 1. Molecular dynamics simulations reveal consistent interaction patterns and stable binding profiles across independent trajectories. For experimental validation, Wistar rats with surgically induced Alzheimer’s disease were orally treated with quercetin (30 mg/kg) for 5 weeks. Histological and immunohistochemical analyses of the hippocampus and subventricular zone evaluated neuronal density and astrocytic activation using glial fibrillary acidic protein and vimentin markers. Computational results supported the multitarget potential of quercetin through stable enzyme interactions. Consistently, in vivo assays demonstrated increased neuronal density and reduced astrocytic marker expression, suggesting a protective modulation of neuroinflammatory processes. These findings highlight quercetin as a promising scaffold for multitarget therapeutic strategies aimed at mitigating the progression of Alzheimer’s disease.
Visual Question Answering (VQA) aims to design systems capable of understanding visual content and answering related questions in natural language. Despite advances, the task remains an open research problem due to dataset biases, limitations in multimodal fusion, and limited interpretability of current models. This paper provides a comprehensive and structured review of VQA research from 2014 to 2025, synthesizing datasets, evaluation protocols, modeling strategies, and emerging trends. In addition, we conduct an empirical proof-of-concept evaluation on the VizWiz dataset to examine how literature-driven design choices influence performance in assistive VQA settings. The experiments employ lightweight multimodal configurations based on pre-trained feature extractors (including BERT, CLIP, and BLIP2), combined via simple feature concatenation, and modeled using a classical MLP. Seven configurations were systematically evaluated using accuracy as the primary metric. The results indicate that informed combinations of established components can yield competitive performance while maintaining low computational complexity. Overall, this work bridges critical synthesis and empirical evaluation, offering an updated synthesis of the field and empirical insights into multimodal design strategies.
Background/Objectives: Alzheimer’s disease (AD) is a progressive, multifactorial neurodegenerative condition characterized by neurofibrillary tangles, neuronal loss, cognitive impairment, and accumulation of β-amyloid plaques. Considering the limitations of current treatments, which present adverse effects and only alleviate symptoms without modifying disease progression, there is an urgent need for new therapeutic approaches. This study aimed to investigate the neuroprotective potential of synthetic derivatives of (1E,4E)-1,5-bis[(het)aryl]penta-1,4-dien-3-ones, focusing on the inhibition of the cholinesterase enzymes acetylcholinesterase (AChE) and butyrylcholinesterase (BChE), targets directly related to the cholinergic deficit observed in AD. Methods: The compounds were initially synthesized by aldol condensation reactions, with subsequent physicochemical characterization. They were then subjected to in silico assays that demonstrated high binding affinity to the active sites of AChE and BChE. The derivatives were evaluated in vitro for their inhibitory activity on these enzymes and in vivo in an experimental model of AD induced by streptozotocin in Wistar rats. Results: The synthesized derivatives showed favorable predicted interactions with the active sites of AChE and BChE, supporting their potential as cholinesterase inhibitors. In vitro assays demonstrated inhibitory activity against both enzymes, with selected derivatives showing improved activity compared with the parent scaffold. In the in vivo model, treatment with the selected compounds was associated with neuroprotective effects, suggesting preservation of nervous tissue integrity under AD-like conditions. Conclusions: These findings indicate that derivatives may represent promising candidates for further investigation as multitarget agents for AD. The study reinforces the relevance of integrating organic synthesis, molecular modeling, enzymatic assays, and in vivo evaluation in the search for new therapeutic strategies for neurodegenerative diseases.