PANDA Platform

PANDA, Platform for Abused-Drugs and Neurological Diseases Association, serves as a portal for accessing the computational resources developed, disseminated and maintained by the CDAR Center. PANDA is a computational hub, currently under construction, that will enable the integrated use of DA chemogenomics knowledgebase (DAKB) with the Center's structural proteome- and genome-scale tools, as well as evaluation of missense variants tools toward establishing the link between molecular, cellular, and systemic aspects of DA and associated disorders.

PANDA is being designed to

enable broad data- and tool-sharing,
provide guidance and task-driven protocols to users in order to facilitate access to, and usage of, advanced computational tools for DA research,
promote new interdisciplinary collaborations and synergistically accelerate existing interactions between computational and experimental communities doing DA-related research,
and thereby advance our understanding of deep associations between DA, neurological diseases and other related disorders, including pain, asthma and psychiatric disorders.

To access different tools, click the arrows on the PANDA wheel.
To view an example protocol that takes advantage of a suite of tools, click here.

This predictor was built by applying the support vector machine (SVM) and LiCABEDS algorithms on four types of fingerprints of 1593 reported compounds.

Chemogenomics Database for Drug abuse Research or Drug Abuse knowledgebase (DA-KB) is designed for facilitating data-sharing and information exchange among scientific research communities for drug abuse, including genes, proteins, small molecules and signal pathways, with online structure search functions and data analysis tools implemented.

The National Center of Excellence for Computational Drug Abuse Research (CDAR), is a joint initiative between the University of Pittsburgh (Pitt) and Carnegie Mellon University (CMU), funded by the NIH National Institute on Drug Abuse (NIDA).

CMM (Coupled Mixed Model) is used to simulnateously conduct genetic analysis for independently collected data sets of different related phenotypes. CMM aims to achieve this goal by inferring all the information enclosed by dashed lines in the following figure.

CS-LMM is used to detect the weaker genetic association conditioned on the stronger validated associations.

DruGUI is a VMD plugin designed for setup and analysis of simulations containing small organic molecules (probes) for druggability assessment. DruGUI can incorporate a diverse set of molecules from CHARMM General Force Field (CGenFF) into simulations (see details).

Tools included in DynOmics are ENM, iGNM, ProDy, ANM and Druggability.

GPCR Prediction Online (DAKB-GPCRs) predicts the BioActivities on 86 drug abuse related GPCRs (G Protein-Coupled Receptors) for a query compound and provides a handy user interface for viewing, downloading and plotting the output results.

GenAMap takes in standard SNP data format (like plink data), or csv/tsv data and outputs visualizations ofIdentified association matrix between SNPs and traits.

HTDocking is designed and constructed to Docknig mulitple drugable protein targets with high-throuput to explore the potiential pharmacology of a small molecule. This server will facility data-sharing and information exchange among scientific research communities with online search functions and data analysis tools implemented for designing bioactive compounds.

Cells are tightly packed with structures and molecules that carry out the day-to-day operations of living. Understanding how cellular design dictates function is essential to understanding life and disease, in the brain, heart, or elsewhere. MCell (Monte Carlo cell) is a program that uses spatially realistic 3-D cellular models and specialized Monte Carlo algorithms to simulate the movements and reactions of molecules within and between cells—cellular microphysiology.

PLAT is a Pathway Learning and Analysis Tool using probabilistic approximation techniques that can analyze the dynamics of biological pathway models using dynamic Bayesian networks (DBN) based approximation techniques. It supports System Biology Mark-up Language (SBML) format and has a GPU-based implementation for high-performance computing.

QuartataWebIdentifies protein-drug interaction by learning a probabilistic latent factor model of drug-target interactions and enabling users to predict the interactions of: any known drug against all targets (by providing only a drugIdentifier), any target against all known, approved drugs (by providing only a targetIdentifier) and any drug-target pair (by providing both a drug and a targetIdentifier).

This tool predicts the pathogenicity for Single Amino acid Variants (SAVs) by employing a Random Forest classifier trained on sequence-, structure- and dynamics-based features.

SMOKE is a Statistical MOdel checKing tool for Estimating unknown parameters of dynamical models. It can utilize both quantitative data and qualitative knowledge for calibrating large models with hundreds of unknown parameters. It was originally developed for analyzing ordinary differential equation (ODE) models of biological networks, and currently being generalized to other modeling formalisms including stochastic models, rule-based models, and hybrid models.

This tool predicts the pathogenicity for Single Amino acid Variants (SAVs) by employing a Random Forest classifier trained on sequence-, structure- and dynamics-based features.

TargetHunter of Small Molecule is designed and constructed toIdentify possible targets of small molecules by searching the available bioactive compound-target pairs reported from literature using the query structure. This server will facility data-sharing and information exchange among UPCMLD scientific research communities with online structure search functions and data analysis tools implemented.