Horacio Pérez-Sánchez

Managing, analysing, and integrating big data in medical bioinformatics: open problems and future perspectives.

The explosion of the data both in the biomedical research and in the healthcare systems demands urgent solutions. In particular, the research in omics sciences is moving from a hypothesis-driven to a data-driven approach. Healthcare is additionally always asking for a tighter integration with biomedical data in order to promote personalized medicine and to provide better treatments. Efficient analysis and interpretation of Big Data opens new avenues to explore molecular biology, new questions to ask about physiological and pathological states, and new ways to answer these open issues. Such analyses lead to better understanding of diseases and development of better and personalized diagnostics and therapeutics. However, such progresses are directly related to the availability of new solutions to deal with this huge amount of information. New paradigms are needed to store and access data, for its annotation and integration and finally for inferring knowledge and making it available to researchers. Bioinformatics can be viewed as the “glue” for all these processes. A clear awareness of present high performance computing (HPC) solutions in bioinformatics, Big Data analysis paradigms for computational biology, and the issues that are still open in the biomedical and healthcare fields represent the starting point to win this challenge.

High-Throughput parallel blind Virtual Screening using BINDSURF

BackgroundVirtual Screening (VS) methods can considerably aid clinical research, predicting how ligands interact with drug targets. Most VS methods suppose a unique binding site for the target, usually derived from the interpretation of the protein crystal structure. However, it has been demonstrated that in many cases, diverse ligands interact with unrelated parts of the target and many VS methods do not take into account this relevant fact.ResultsWe present BINDSURF, a novel VS methodology that scans the whole protein surface in order to find new hotspots, where ligands might potentially interact with, and which is implemented in last generation massively parallel GPU hardware, allowing fast processing of large ligand databases.ConclusionsBINDSURF is an efficient and fast blind methodology for the determination of protein binding sites depending on the ligand, that uses the massively parallel architecture of GPUs for fast pre-screening of large ligand databases. Its results can also guide posterior application of more detailed VS methods in concrete binding sites of proteins, and its utilization can aid in drug discovery, design, repurposing and therefore help considerably in clinical research.

Effect of inclusion of hydroxycinnamic and chlorogenic acids from green coffee bean in β-cyclodextrin on their interactions with whey, egg white and soy protein isolates

The aim of the study was to characterise the interactions of hydroxycinnamic and chlorogenic acids (CHAs) from green coffee, with isolates of proteins from egg white (EWP), whey (WPC) and soy (SPI), depending on pH and temperature. The binding degree was determined by liquid chromatography coupled to a diode array detector and an ultrahigh resolution hybrid quadruple-time-of-flight mass spectrometer with ESI source (LC-QTOF-MS/MS). As a result of binding, the concentration of CHAs in proteins ranged from 9.44-12.2, 11.8-13.1 and 12.1-14.4g/100g for SPI, WPC and EWP, respectively. Thermodynamic parameters of protein-ligand interactions were determined by isothermal titration calorimetry (ITC) and energetics of interactions at the atomic level by molecular modelling. The amount of CHAs released during proteolytic digestion was in the range 0.33-2.67g/100g. Inclusion of CHAs with β-cyclodextrin strongly limited these interactions to a level of 0.03-0.06g/100g.

In silico discovery of a compound with nanomolar affinity to antithrombin causing partial activation and increased heparin affinity.

The medical and socioeconomic relevance of thromboembolic disorders promotes an ongoing effort to develop new anticoagulants. Heparin is widely used as activator of antithrombin but incurs side effects. We screened a large database in silico to find alternative molecules and predicted d-myo-inositol 3,4,5,6-tetrakisphosphate (TMI) to strongly interact with antithrombin. Isothermal titration calorimetry confirmed a TMI affinity of 45 nM, higher than the heparin affinity (273 nM). Functional studies, fluorescence analysis, and citrullination experiments revealed that TMI induced a partial activation of antithrombin that facilitated the interaction with heparin and low affinity heparins. TMI improved antithrombin inhibitory function of plasma from homozygous patients with antithrombin deficiency with a heparin binding defect and also in a model with endothelial cells. Our in silico screen identified a new, non-polysaccharide scaffold able to interact with the heparin binding domain of antithrombin. The functional consequences of this interaction were experimentally characterized and suggest potential anticoagulant therapeutic applications.

Optimization methods for virtual screening on novel computational architectures.

The numerous virtual screening (VS) methods that are used today in drug discovery processes differ mainly by the way they model the receptor and/or ligand and by the approach to perform screening. All these methods have in common that they screen databases of chemical compounds containing up to millions of ligands i.e. ZINC database. Larger databases increase the chances of generating hits or leads, but the computational time needed for the calculations increases not only with the size of the database but also with the accuracy of the VS method and the model. Fast docking methods with atomic resolution require a few minutes per ligand, while molecular dynamics-based approaches still require hundreds or thousands of hours per ligand. Therefore, the limitations of VS predictions are directly related to a lack of computational resources, a major bottleneck that prevents the application of detailed, high-accuracy models to VS. The current increase in available computer power at low cost due to novel computational architectures would enhance considerably the performance of the different VS methods and the quality and quantity of the conclusions we can get from screening. In this review, we will discuss recent trends in modeling techniques which, in combination with novel hardware platforms, yield order-of-magnitude improvements in the processing speeds of VS methods. We show the state of the art of VS methods as applied with novel computational architectures and the current trends of advanced computing.

Cuminaldehyde as the Major Component of Cuminum cyminum, a Natural Aldehyde with Inhibitory Effect on Alpha-Synuclein Fibrillation and Cytotoxicity.

Fibrillation of alpha-synuclein (α-SN) is a critical process in the pathophysiology of several neurodegenerative diseases, especially Parkinsons disease. Application of bioactive inhibitory compounds from herbal extracts is a potential therapeutic approach for this cytotoxic process. Here, we investigated the inhibitory effects of the Iranian Cuminum cyminum essential oil on the fibrillation of α-SN. Analysis of different fractions from the total extract identified cuminaldehyde as the active compound involved in the antifibrillation activity. In comparison with baicalein, a well-known inhibitor of α-SN fibrillation, cuminaldehyde showed the same activity in some aspects and a different activity on other parameters influencing α-SN fibrillation. The presence of spermidine, an α-SN fibrillation inducer, dominantly enforced the inhibitory effects of cuminaldehyde even more intensively than baicalein. Furthermore, the results from experiments using preformed fibrils and monobromobimane-labeled monomeric protein also suggest that cuminaldehyde prevents α-SN fibrillation even in the presence of seeds, having no disaggregating impact on the preformed fibrils. Structural studies showed that cuminaldehyde stalls protein assembly into β-structural fibrils, which might be achieved by the interaction with amine groups through its aldehyde group as a Schiff base reaction. This assumption was supported by FITC labeling efficiency assay. In addition, cytotoxicity assays on PC12 cells showed that cuminaldehyde is a nontoxic compound, treatment with cuminaldehyde throughout α-SN fibrillation showed no toxic effects on the cells. Taken together, these results show for the first time that the small abundant natural compound, cuminaldehyde, can modulate α-SN fibrillation. Hence, suggesting that such natural active aldehyde could have potential therapeutic applications.

Effective Parallelization of Non-bonded Interactions Kernel for Virtual Screening on GPUs

In this work we discuss the benefits of using massively parallel architectures for the optimization of Virtual Screeningmethods.We empirically demonstrate that GPUs are well suited architecture for the acceleration of non-bonded interaction kernels, obtaining up to a 260 times sustained speedup compared to its sequential counterpart version.

The Inhibitory Effects of Bioactive Compounds of Tomato Juice Binding to Hepatic HMGCR: In Vivo Study and Molecular Modelling

The hypocholesterolemic effect of tomato juice has been investigated in an intervention study with rats, along with the possible inhibition effect of bioactive tomato compounds binding to the HMGCR enzyme. Two experimental groups (n = 8 Sprague-Dawley rats) were fed ad libitum for five weeks, with water or tomato juice provided to the control and intervention groups, respectively. Total, LDL and HDL cholesterol, and total triglycerides were analysed in plasma, and the lycopene content and the expression and activity of the enzyme HMGCR were determined in liver samples. A computational molecular modelling was carried out to determine the interactions between HMGCR and lycopene, chlorogenic acid and naringenin. Total, LDL and HDL cholesterol were significantly lower in the intervention group after the intake of tomato juice. In addition, a significant reduction in HMGCR activity was observed, although this was not accompanied by changes in gene expression. The molecular modelling showed that components of tomato can bind to the active site of the enzyme and compete with the ligand HMGCoA. Lycopene, from tomato juice, accumulates in the liver and can inhibit the activity of the rate-limiting enzyme of cholesterol biosynthesis, HMGCR.

Labelling Herceptin with a novel oxaliplatin derivative: a computational approach towards the selective drug delivery

AbstractThe clinical use of platinum(II)-based drugs has serious side effects due to the non-specific reactions with both malignant and normal cells. To circumvent such major drawback, novel metallodrugs might be combined with suitable carrier molecules, as antibodies, to ensure selective attacks on tumours while sparing healthy tissues. In this contribution, we investigate the stability of a novel oxaliplatin derivate drug embedded in Herceptin (trastuzumab), an antibody which is able to recognise breast cancer cells, by using a wide panel of theoretical tools: docking, molecular dynamics and quantum calculations. Our calculations reveal the binding mechanism: the drug initially interacts non-covalently with the Pro40A and Asp167A residues, and the nitrogen of His171B subsequently replaces one of the water molecules coordinated to the platinum center, where the latter step reversibly fixes the drug into the antibody. These data might be used to further rationalise the synthesis of improved drugs beyond classical platinum(II) derivatives by improving the ligand-protein coupling mode. Graphical AbstractA wide panel of theoretical tools is used to determine the chemical interactions stablished between a novel platinum(II)-based drug when linked to the Herceptin antibody.

Structure-based discovery of clinically approved drugs as Zika virus NS2B-NS3 protease inhibitors that potently inhibit Zika virus infection in vitro and in vivo

&NA; Zika virus (ZIKV) infection may be associated with severe complications in fetuses and adults, but treatment options are limited. We performed an in silico structure‐based screening of a large chemical library to identify potential ZIKV NS2B‐NS3 protease inhibitors. Clinically approved drugs belonging to different drug classes were selected among the 100 primary hit compounds with the highest predicted binding affinities to ZIKV NS2B‐NS3‐protease for validation studies. ZIKV NS2B‐NS3 protease inhibitory activity was validated in most of the selected drugs and in vitro anti‐ZIKV activity was identified in two of them (novobiocin and lopinavir‐ritonavir). Molecular docking and molecular dynamics simulations predicted that novobiocin bound to ZIKV NS2B‐NS3‐protease with high stability. Dexamethasone‐immunosuppressed mice with disseminated ZIKV infection and novobiocin treatment had significantly (P < 0.05) higher survival rate (100% vs 0%), lower mean blood and tissue viral loads, and less severe histopathological changes than untreated controls. This structure‐based drug discovery platform should facilitate the identification of additional enzyme inhibitors of ZIKV.