Search

Abstract Drug design is an important area of study for pharmaceutical businesses. However, low efficacy, off-target delivery, time consumption, and high cost are challenges and can create barriers that impact this process. Deep Learning models are emerging as a promising solution to perform de novo drug design, i.e., to generate drug-like molecules tailored to specific needs. However, stereochemistry was not explicitly considered in the generated molecules, which is inevitable in targeted-oriented molecules. This paper proposes a framework based on Feedback Generative Adversarial Network (GAN) that includes optimization strategy by incorporating Encoder–Decoder, GAN, and Predictor deep models interconnected with a feedback loop. The Encoder–Decoder converts the string notations of molecules into latent space vectors, effectively creating a new type of molecular representation. At the same time, the GAN can learn and replicate the training data distribution and, therefore, generate new compounds. The feedback loop is designed to incorporate and evaluate the generated molecules according to the multiobjective desired property at every epoch of training to ensure a steady shift of the generated distribution towards the space of the targeted properties. Moreover, to develop a more precise set of molecules, we also incorporate a multiobjective optimization selection technique based on a non-dominated sorting genetic algorithm. The results demonstrate that the proposed framework can generate realistic, novel molecules that span the chemical space. The proposed Encoder–Decoder model correctly reconstructs 99% of the datasets, including stereochemical information. The model’s ability to find uncharted regions of the chemical space was successfully shown by optimizing the unbiased GAN to generate molecules with a high binding affinity to the Kappa Opioid and Adenosine $$A_ {2a}$$ A 2 a receptor. Furthermore, the generated compounds exhibit high internal and external diversity levels 0.88 and 0.94, respectively, and uniqueness.

Abstract Background Several computational advances have been achieved in the drug discovery field, promoting the identification of novel drug–target interactions and new leads. However, most of these methodologies have been overlooking the importance of providing explanations to the decision-making process of deep learning architectures. In this research study, we explore the reliability of convolutional neural networks (CNNs) at identifying relevant regions for binding, specifically binding sites and motifs, and the significance of the deep representations extracted by providing explanations to the model’s decisions based on the identification of the input regions that contributed the most to the prediction. We make use of an end-to-end deep learning architecture to predict binding affinity, where CNNs are exploited in their capacity to automatically identify and extract discriminating deep representations from 1D sequential and structural data. Results The results demonstrate the effectiveness of the deep representations extracted from CNNs in the prediction of drug–target interactions. CNNs were found to identify and extract features from regions relevant for the interaction, where the weight associated with these spots was in the range of those with the highest positive influence given by the CNNs in the prediction. The end-to-end deep learning model achieved the highest performance both in the prediction of the binding affinity and on the ability to correctly distinguish the interaction strength rank order when compared to baseline approaches. Conclusions This research study validates the potential applicability of an end-to-end deep learning architecture in the context of drug discovery beyond the confined space of proteins and ligands with determined 3D structure. Furthermore, it shows the reliability of the deep representations extracted from the CNNs by providing explainability to the decision-making process.

The discovery of potential Drug-Target Interactions (DTIs) is a determining step in the drug discovery and repositioning process, as the effectiveness of the currently available antibiotic treatment is declining. Although putting efforts on the traditional in vivo or in vitro methods, pharmaceutical financial investment has been reduced over the years. Therefore, establishing effective computational methods is decisive to find new leads in a reasonable amount of time. Successful approaches have been presented to solve this problem but seldom protein sequences and structured data are used together. In this paper, we present a deep learning architecture model, which exploits the particular ability of Convolutional Neural Networks (CNNs) to obtain 1D representations from protein sequences (amino acid sequence) and compounds SMILES (Simplified Molecular Input Line Entry System) strings. These representations can be interpreted as features that express local dependencies or patterns that can then be used in a Fully Connected Neural Network (FCNN), acting as a binary classifier. The results achieved demonstrate that using CNNs to obtain representations of the data, instead of the traditional descriptors, lead to improved performance. The proposed end-to-end deep learning method outperformed traditional machine learning approaches in the correct classification of both positive and negative interactions. [ABSTRACT FROM AUTHOR]

A fundamental experimental investigation is described of the structure and streamwise development of a confluent boundary layer generated by the interaction of a leading-edge slat wake with the turbulent boundary layer on the main element of a multi-element airfoil model. The slat and airfoil model geometry are both fully two dimensional. The research is performed in an attempt to investigate the flow physics of confluent boundary layers and to build an archival database on the interaction of the slat wake and the main element wall layer. In addition, an attempt is made to identify clearly the role that slat wake/airfoil boundary-layer confluence has on lift production and how this occurs. Although complete laser Doppler velocimetry flow surveys were performed for a variety of slat gap and overhang settings, the focus is on two cases representing both strong and weak wake boundary-layer confluence. For those interested in performing numerical simulations, the full database is available from the authors.

Objectives: The aim of this study was to compare the intraindividual effects of contrast material with two different iodine concentrations on the conspicuity of hepatocellular carcinoma (HCC) and vascular and hepatic contrast enhancement during multiphasic, 64-section multidetector row CT (MDCT) in patients with cirrhosis using two contrast medium injection techniques. Methods: Patients were randomly assigned to one of two groups with an equal iodine dose but different contrast material injection techniques: scheme A, fixed injection duration (25 s), and scheme B, fixed injection flow rate (4 mls-1). For each group, patients were randomised to receive both moderate-concentration contrast medium (MCCM) and high-concentration contrast medium (HCCM) during two CT examinations within 3 months. Enhancement of the aorta, liver and portal vein and the tumour-to- liver contrast-to-noise ratio (CNR) were compared between MCCM and HCCM. Results: 30 patients (mean age 59 years; range 45-80 years; 16 patients in scheme A and 14 in scheme B) with a total of 31 confirmed HCC nodules were prospectively enrolled. For scheme B, the mean contrast enhancement of the aorta and tumour-to-liver CNR were significantly higher with HCCM than with MCCM during the hepatic arterial phase (+350.5 HU vs +301.1 HU, p=0.00l, and +7.5 HU vs +5.5 HU, p=0.004). For both groups, there was no significant difference between MCCM and HCCM for all other comparisons. Conclusion: For a constant injection flow rate, HCCM significantly improves the conspicuity of HCC lesions and aortic enhancement during the hepatic arterial phase on 64-section MDCT in patients with cirrhosis. [ABSTRACT FROM AUTHOR]

Xenotransplantation of porcine tissue to human recipients promises to alleviate the organ shortage. Human antibody-mediated and cell-mediated immune responses against porcine grafts, however, represent barriers to successful xenotransplantation. We compared neonatal porcine islet cells (NPICs) and neonatal porcine splenocytes for the ability to stimulate proliferation of human peripheral blood lymphocytes (PBLs), and for their susceptibility to human natural killer (NK) and cytotoxic T-lymphocyte (CTL)-mediated lysis. Human peripheral blood CD4+ lymphocytes showed strong proliferation in response to NPICs, likely because of occasional swine leukocyte antigen (SLA) class II+ cells in the NPIC preparations. In contrast, human peripheral blood CD8+ lymphocytes did not proliferate in response to NPICs, although they showed clear responses to both porcine splenocytes and endothelial cells. Both human CTL-raised-against-porcine splenocytes and endogenous NK cells lysed porcine splenocytes, but the same cells showed little or no lytic activity against NPICs. Lysis of porcine splenocyte targets was completely abrogated by pretreatment of the human NK or CTL populations with concana-mycin A, suggesting a perforin-dependent effector mechanism. Pretreatment of the NPIC targets with proinflammatory porcine cytokines to upregulate SLA class I expression failed to enhance human CTL-mediated lysis. However, lysis of NPICs by human CTLs could be elicited when a lectin was added to form stable effector:target cell conjugates. It appears that NPICs do not express sufficiently high levels of co-stimulatory and/or adhesion molecules to either activate human CD8+ T-cells or to be effective targets for activated human CTLs. These data suggest that NPICs may not be destroyed by NK- or CTL-mediated lytic mechanisms after transplantation into humans. [ABSTRACT FROM AUTHOR]

Forty-three patients who were scheduled to undergo a percutaneous liver biopsy were evaluated with Doppler sonography to determine the hepatic arterial resistive index (RI). The histologic specimens were graded by a pathologist regarding cirrhosis and inflammation. The specimens demonstrated no cirrhosis in 12 of 43 (28%) patients, early cirrhosis in 10 of 43 (23%), and established cirrhosis in 21 of 43 (49%). Analysis also revealed that inflammation was absent in three of 43 (7%) patients, minimal in seven of 43 (16%), mild in 17 of 43 (40%), moderate in 13 of 43 (30%), and severe in three of 43 (7%). Hepatic artery RIs (without correction for heart rate) ranged from 0.64 +/- 0.06 in patients with early cirrhosis to 0.68 +/- 0.09 in patients with severe inflammation. There was no significant correlation between the degree of cirrhosis and/or inflammation and hepatic artery RI (with or without correction for heart rate). We conclude that Doppler determination of hepatic artery RIs is not a reliable method of predicting the severity of hepatic cirrhosis and/or inflammation. [ABSTRACT FROM AUTHOR]