Drug design is an important area of study for pharmaceutical businesses. However, low efficacy, off-target delivery, time consumption, and high cost are challenges can create barriers that impact this process. Deep Learning models emerging as a promising solution to perform de novo drug design, i.e., generate drug-like molecules tailored specific needs. stereochemistry was not explicitly considered in the generated molecules, which inevitable targeted-oriented molecules. This paper proposes...

Molecular property prediction is a critical step in drug discovery. Deep learning (DL) has accelerated the discovery of compounds with desirable molecular properties for successful development. However, low-data problem which makes it hard to solve by regular DL approaches. Graph neural networks (GNNs) operate on graph-structured data using neighborhood aggregation facilitate properties. Nonetheless, GNNs struggle model global-semantic structure graph embeddings prediction. Recently,...

Codon usage and codon-pair context are important gene primary structure features that influence mRNA decoding fidelity. In order to identify general rules shape minimize error, we have carried out a large scale comparative analysis of 119 fully sequenced genomes.We developed mathematical software tools for analysis. These methodologies unveiled species specific govern evolution mRNAs in the 3 domains life. We show bacterial archeal is mainly dependent on constraints imposed by translational...

Abstract In this work, we explore the potential of deep learning to streamline process identifying new drugs through computational generation molecules with interesting biological properties. Two neural networks compose our targeted framework: Generator , which is trained learn building rules valid employing SMILES strings notation, and Predictor evaluates newly generated compounds by predicting their affinity for desired target. Then, optimized Reinforcement Learning produce bespoken The...

The accurate identification of Drug-Target Interactions (DTIs) remains a critical turning point in drug discovery and understanding the binding process. Despite recent advances computational solutions to overcome challenges vitro vivo experiments, most proposed silico-based methods still focus on binary classification, overlooking importance characterizing DTIs with unbiased strength values properly distinguish primary interactions from those off-targets. Moreover, several these usually...

Abstract Background MicroRNAs (miRNAs) are a new class of small RNAs approximately 22 nucleotides in length that control eukaryotic gene expression by fine tuning mRNA translation. They regulate wide variety biological processes, namely developmental timing, cell differentiation, proliferation, immune response and infection. For this reason, their identification is essential to understand biology. Their size, low abundance high instability complicated early identification, however...

The discovery of potential Drug-Target Interactions (DTIs) is a determining step in the drug and repositioning process, as effectiveness currently available antibiotic treatment declining. Although putting efforts on traditional <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">in vivo</i> or vitro</i> methods, pharmaceutical financial investment has been reduced over years. Therefore, establishing effective computational methods decisive to...

Molecular property prediction is an essential task in drug discovery. Recently, deep neural networks have accelerated the discovery of compounds with improved molecular profiles for effective development. In particular, graph (GNNs) played a pivotal role identifying promising candidates desirable properties. However, it common only few molecules to share same set properties, which presents low-data problem unanswered by regular machine learning (ML) approaches. Transformer also emerged as...

Due to the constant increase in cancer rates, disease has become a leading cause of death worldwide, enhancing need for its detection and treatment. In era personalized medicine, main goal is incorporate individual variability order choose more precisely which therapy prevention strategies suit each person. However, predicting sensitivity tumors anticancer treatments remains challenge. this work, we propose two deep neural network models predict impact drugs through half-maximal inhibitory...

Background Mild cognitive impairment (MCI) exhibits considerable heterogeneity, requiring accurate characterization through classification and prognostic models. In clinical research, data-driven models offer valuable insights for classification, stratification, predicting progression to dementia. Objective We implemented computational techniques characterize MCI patients develop multistate Alzheimer's disease (AD). Methods Datasets comprising 544 from Coimbra University Hospital 497 the...

De novo experimental drug discovery is an expensive and time-consuming task. It requires the identification of drug-target interactions (DTIs) towards targets biological interest, either to inhibit or enhance a specific molecular function. Dedicated computational models for protein simulation DTI prediction are crucial speed reduce costs associated with identification. In this paper we present pipeline that enables putative leads repositioning can be applied any microbial proteome, as long...

The design of compounds that target specific biological functions with relevant selectivity is critical in the context drug discovery, especially due to polypharmacological nature most existing molecules. In recent years, silico-based methods combined deep learning have shown promising results de novo challenge, leading potential leads for biologically interesting targets. However, several these overlook importance certain properties, such as validity rate and selectivity, or simplify...

The proper assessment of target-specific compound selectivity is paramount in the drug discovery context, promoting identification drug-target interactions (DTIs) and potential leads. On that account, accurate prediction an unbiased binding affinity (DTA) metric pivotal to understanding process. Most silico computational approaches, however, neglect inter-dependency proteomics, chemical, pharmacological spaces explainability during model construction. Furthermore, these methods have yet...

Abstract Background The oral cavity is a complex ecosystem where human chemical compounds coexist with particular microbiota. However, shifts in the normal composition of this microbiota may result onset ailments, such as periodontitis and dental caries. In addition, it known that microbial colonization mediated by protein-protein interactions (PPIs) between host microorganisms. Nevertheless, kind PPIs still largely undisclosed. To elucidate these interactions, we have created computational...

Abstract Background Advances in biotechnology and high-throughput methods for gene analysis have contributed to an exponential increase the number of scientific publications these fields study. While much data results described articles are entered annotated various existing biomedical databases, literature is still major source information. There is, therefore, a growing need text mining information retrieval tools help researchers find relevant their To tackle this, several been proposed...

Identifying ZIKV factors interfering with human host pathways represents a major challenge in understanding tropism and pathogenesis. The integration of proteomic, gene expression Protein-Protein Interactions (PPIs) established between proteins predicted by the OralInt algorithm identified 1898 interactions medium or high score (≥0.7). Targets implicated vesicular traffic docking were identified. New receptors involved endocytosis as entry targets, using both clathrin-dependent (17...

The process of placing new drugs into the market is time-consuming, expensive and complex. application computational methods for designing molecules with bespoke properties can contribute to saving resources throughout this process. However, fundamental be optimized are often not considered or conflicting each other. In work, we propose a novel approach consider both biological property bioavailability compounds through deep reinforcement learning framework targeted generation compounds. We...

Abstract The generation of candidate hit molecules with the potential to be used in cancer treatment is a challenging task. In this context, computational methods based on deep learning have been employed improve silico drug design methodologies. Nonetheless, applied strategies focused solely chemical aspect compounds, disregarding likely biological consequences for organism’s dynamics. Herein, we propose method implement targeted molecular that employs information, namely,...

Deep learning has gained major popularity in automated feature extraction from images, audio and text. We present two case studies where deep can have a key impact. The first study consists of graphic logo detection based on fast region-based convolutional networks (FRCN). This method tackles the issue different size positioning by looking for scale invariant regions. avoids full image search while improving overall object detection. Furthermore, instead building neural (CNN) scratch,...