RESUMEN
Identification of DNA-binding proteins is essential in studying cellular activities as the DNA-binding proteins play a pivotal role in gene regulation. In this study, we propose newDNA-Prot, a DNA-binding protein predictor that employs support vector machine classifier and a comprehensive feature representation. The sequence representation are categorized into 6 groups: primary sequence based, evolutionary profile based, predicted secondary structure based, predicted relative solvent accessibility based, physicochemical property based and biological function based features. The mRMR, wrapper and two-stage feature selection methods are employed for removing irrelevant features and reducing redundant features. Experiments demonstrate that the two-stage method performs better than the mRMR and wrapper methods. We also perform a statistical analysis on the selected features and results show that more than 95% of the selected features are statistically significant and they cover all 6 feature groups. The newDNA-Prot method is compared with several state of the art algorithms, including iDNA-Prot, DNAbinder and DNA-Prot. The results demonstrate that newDNA-Prot method outperforms the iDNA-Prot, DNAbinder and DNA-Prot methods. More specific, newDNA-Prot improves the runner-up method, DNA-Prot for around 10% on several evaluation measures. The proposed newDNA-Prot method is available at http://sourceforge.net/projects/newdnaprot/
Asunto(s)
Proteínas de Unión al ADN , Máquina de Vectores de Soporte , Secuencia de Aminoácidos , Proteínas de Unión al ADN/química , Proteínas de Unión al ADN/metabolismo , Bases de Datos de Proteínas , Estructura Secundaria de ProteínaRESUMEN
The protein quaternary structure is very important to the biological process. Predicting their attributes is an essential task in computational biology for the advancement of the proteomics. However, the existing methods did not consider sufficient properties of amino acid. To end this, we proposed a hybrid method Quad-PRE to predict protein quaternary structure attributes using the properties of amino acid, predicted secondary structure, predicted relative solvent accessibility, and position-specific scoring matrix profiles and motifs. Empirical evaluation on independent dataset shows that Quad-PRE achieved higher overall accuracy 81.7%, especially higher accuracy 92.8%, 93.3%, and 90.6% on discrimination for trimer, hexamer, and octamer, respectively. Our model also reveals that six features sets are all important to the prediction, and a hybrid method is an optimal strategy by now. The results indicate that the proposed method can classify protein quaternary structure attributes effectively.
Asunto(s)
Aminoácidos/química , Biología Computacional/métodos , Estructura Cuaternaria de Proteína , Proteínas/química , Área Bajo la Curva , Bases de Datos de Proteínas , Humanos , Proteómica/métodos , Curva ROC , Reproducibilidad de los Resultados , Programas Informáticos , Máquina de Vectores de SoporteRESUMEN
The RNA polymerase of Influenza A virus (IAV), which is comprised of three units PA, PB1 and PB2, is involved in transcription and replication of the influenza virus. In order to develop effective treatment for IAV, researchers have focused on designing drugs targeting IAV polymerase. Currently, crystal structures of the IAV polymerase PA-PB1, PB1-PB2 complexes and the PA subunit have been obtained by several groups, providing useful information regarding potential binding sites in drug design. However, to gain full understanding of the molecular mechanism of IAV polymerase in viral transcription and replication, thereby aiding drug development, a complete atomistic structure of the RNA polymerase is required. In this paper, we employed computer-aided drug design tools to describe the complete structure of the RNA polymerase and proposed a putative mechanism. We predict that the combination of Vancomycin and Oseltamivir will be an effective drug to universally treat IAVs with no resultant drug resistance if this putative mechanism is true.
Asunto(s)
Antivirales/farmacología , Diseño Asistido por Computadora , Diseño de Fármacos , Terapia Molecular Dirigida , Antivirales/administración & dosificación , Sitios de Unión , ARN Polimerasas Dirigidas por ADN/química , ARN Polimerasas Dirigidas por ADN/efectos de los fármacos , ARN Polimerasas Dirigidas por ADN/metabolismo , Farmacorresistencia Viral , Quimioterapia Combinada , Humanos , Virus de la Influenza A/efectos de los fármacos , Virus de la Influenza A/enzimología , Gripe Humana/tratamiento farmacológico , Gripe Humana/virología , Oseltamivir/administración & dosificación , Oseltamivir/farmacología , Vancomicina/administración & dosificación , Vancomicina/farmacología , Replicación Viral/efectos de los fármacosRESUMEN
BACKGROUND: The H1N1 pandemic in 2009 and the H5N1 pandemic in 2005 demonstrated that the drugs approved to treat influenza A viruses have low efficacy. This provided a stimulus for new studies of influenza A viruses in the context of the methods used in drug design developed over the past 100 years. Finding new universal drugs is the ultimate goal but its long time horizon is incompatible with emergency situations created by reoccurring influenza outbreaks. Therefore, we propose a computer-aided method for finding efficacious drugs and drug complexes based on the use of the DrugBank database. METHODS: (1) We start by assembling a panel of target proteins. (2) We then assemble a panel of drugs. (3) This is followed by a selection of benchmark binding pockets based on the panel of target proteins and the panel of drugs. (4) We generate a set of computational features, which measure the efficacy of a drug. (5) We propose a universal program to search for drugs and drug complexes. (6) A case study we report here illustrates how to use this universal program for finding an optimal drug and a drug complex for a given target. (7) Validation of the Azirchromycin and Aspirin complex is provided mathematically. (8) Finally, we propose a simple strategy to validate our computational prediction that the Azirchromycin and Aspirin complex should prove clinically effective. RESULT: A set of computable features are mined and then based on these features, a universal program for finding the potential drug &drug complexes is proposed. Using this universal program, the Azirchromycin and Aspirin complex is selected and its efficacy is predicted mathematically. For clinical validation of this finding, future work is still required.