Pfizer Deploys Intranet-Based Statserver Technology to Improve FDA Reporting

Business Challenge

Pfizer is devoted to discovering, developing, manufacturing, and marketing quality pharmaceutical products. A pharmaceutical leader, the company manufactures prescription drugs including Lipitor® for cardiovascular disease and Accupril® for hypertension and congestive heart failure.

Each year, the global pharmaceutical industry spends $40.1 billion on drug discovery, development and approval. Data analysis software can help researchers to analyze and translate their results into easy-to-read documents for Federal Drug Administration (FDA) approval. Professional reporting with informative, precise graphics translating results can help FDA decision makers to bring safe, effective drugs to market sooner.

Drug discovery and development requires accurate statistical analysis from initial experimental design to final delivery of a drug to a pharmacy. At Pfizer, researchers in its Pharmaceutical Delivery System (PDS) work closely with drug discovery, chemical development and manufacturing associates to bring the best pharmaceutical products to market. This team tests new products for stability, performance and degradation and provides results and recommended expiration dates in detailed reports submitted to the FDA.

At Pfizer, S-PLUS was deployed using StatServer software and a company-wide Intranet for data analysis and FDA reporting. “Historically, our company submitted hundreds of lengthy reports including 30 to 40 pages of graphs to the FDA. Each researcher was responsible for preparing their own reports and as a company we did not have access to standardized analytics. We were looking for a leading software solution that could deliver the powerful analytics and graphic capabilities we needed to our employee’s desktops,” says John Twist, Section Director, PDS Planning and Statistics. “We selected StatServer because it provided its cutting-edge analytics, browser compatibility and integration with complementary Microsoft Office reporting packages.”

Business Solution

Twist was interested in streamlining his team’s data analysis and reporting process. He created a Web-based application that allowed researchers to upload their data into customized analytic applications from a variety of common sources including, most notably, Excel spreadsheets. The data could then be analyzed using cutting-edge statistical techniques, like fitting analyses of covariance (ANCOVA) models with corresponding confidence bands for mean or individual values, offered in the S-PLUS software package. The results were then translated into conditioned multipanel (Trellis) plots for submission to the FDA.

“In the past we needed to plot many variables in several different graphs. Usually, the software allowed us to have one plot per page. Graphics in standard FDA reports often were 30 to 40 pages. Using S-PLUS we were able to provide FDA decision-makers with valuable information on one page. Our new reports provide FDA reviewers with the information they need quickly to make informed decisions,” says Twist. “S-PLUS has allowed each researcher to access standardized analytics and report tools from their desktops improving our reporting process,” says Twist.

“We selected S-PLUS software because it provided us with graphic capabilities and powerful data analysis tools we needed to perform our work,” says Twist. Now, researchers can prepare reports that are consistent, easy-to-read and translate results effectively. Better reports can help bring safe, effective drugs to market faster improving the quality of human lives.

Benefits

Powerful Data Analysis
Efficient Visualization Capabilities
Seamless Integration With Microsoft Office And Other Business Tools for FDA Reporting
StatServer/S-PLUS

Next-Generation Data Analysis Software Aids Faster Drug Development

Researchers studying variability in drug concentrations in patient populations use S-PLUS software to accelerate data analysis to improve the understanding of drug safety and efficacy.

Faster and better drug development can reduce the time taken to make safe and effective drugs available to those who need them most. The pharmaceutical scientists in academia, the pharmaceutical industry, and regulatory agencies such as the Food and Drug Administration (FDA) have been greatly interested in the pharmacometric techniques to understand drug safety and efficacy.

Pharmacometrics is a specialty within clinical pharmacology that addresses the development and application of statistical and mathematical techniques to assess the relationship between drug dose and concentration [pharmacokinetics (PK)], PK and the effect of a drug on the body [pharmacodynamics (PD)], PK and safety/efficacy, and the efficient implementation of model-based strategies in drug development. Thus, pharmacometricians use mathematical and statistical techniques in PK/PD to promote rational drug development with rational pharmacotherapy (right dose for the right patient) as the end product).

Drug discovery and development can be time consuming and expensive. Each year, the global pharmaceutical industry spends at least $10 billion on clinical trials. To counter this expense, next-generation statistical software provides pharmaceutical scientists with the tools they need to discover hidden knowledge buried in data collected during clinical trials. The use of software tools with powerful graphical and modeling capabilities, such as S-PLUS, result in better and faster drug development. The S-PLUS software is based on S-language, which is specifically designed for data visualization and exploration, statistical modeling and programming with data.

Population pharmacokinetics is the study of the sources and correlates of variability in plasma drug concentrations among individuals who are preferably the target patient population receiving clinically relevant doses of a drug of interest. Patient demographic features or pathophysiological variables can be helpful in explaining alterations in the dose-concentration relationship. Population pharmacokinetics applies an exploratory approach to drug development. Integrated information on pharmacokinetics from concentration to time data collected during clinical trials can be gathered using pharmacokinetic techniques. Statistical tools allow the pharmaceutical scientist to explore and discover important hidden knowledge in clinical data sets efficiently.

Exploratory Data Analysis

Using graphical and modern statistical techniques to explore clinical data can isolate and reveal patterns and features in the data. Both of these techniques can uncover unexpected departures from hypothesized models (e.g., departure from a normal distribution). An important element of this exploratory approach is the software’s flexibility to tailor the analysis to the data structure and to respond to patterns uncovered by successive analysis. Exploratory data analysis (EDA) is essential to population pharmacokinetic modeling, and new statistical software tools allow the data analyst to visualize and prototype data more efficiently.

According to Dr. Ene Ette, a leading pharmaceutical scientist with Vertex Pharmaceuticals, Inc., raw computer power, combined with sophisticated data analysis tools like S-PLUS, brings valuable findings to the drug development team more quickly with safety and efficacy in mind. “One of the difficult tasks for a pharmacometrician is to convey findings from pharmacometric analyses to clinicians and other members of the drug development team. Failure to communicate these findings successfully can hamper drug research and development. S-PLUS offers easy-to-read, attractive graphics (Trellis graphics) that can be used to communicate results effectively to the drug development team facilitating the development process,” says Dr. Ette.

Dr. Ette heads a research team that integrates the population pharmacokinetics approach in drug development. Using state-of-the-art statistical software, his team captures, stores, and analyzes population pharmacokinetic and pharmacodynamic data sets to discover hidden knowledge about drugs in development. “We believe that data analysis tools can enhance our capability to bring safe, effective drugs to the market sooner. S-PLUS provides the sophisticated statistical techniques we need to discover patterns and trends in clinical trial data sets and explore meaningful trends in the data we collect. With such information, better and smarter trials can be designed,’ Dr. Ette remarks.

Dr. Ette’s work involves using modern statistical techniques, such as those implemented in S-PLUS, to extract information hidden in population pharmacokinetic and pharmacodynamic data sets. “We selected S-PLUS because it offered powerful graphics combined with state-of-the art statistical techniques that would enable us make the most of the data. The package allowed us to effectively communicate the results to the drug development team,” Dr. Ette remarks. He believes state-of-the-art data analysis software helps researchers to explore, isolate and analyze patterns in population pharmacokinetic and pharmacodynamic data sets, thereby aiding the discovery of information hidden in the data.

Most population analysis procedures are based on explicit data assumptions. The validity of data analysis depends upon the validity of the assumptions. Exploratory data analysis provides powerful diagnostic tools for confirming assumptions or, when the assumptions are not met, for suggesting corrective actions.

Acceleration of Population Pharmacokinetic Analysis

Population pharmacokinetic analysis is time consuming, and software tools can reduce the time spent exploring and analyzing data.

Analytic results are important for designing studies and for filing new drug applications (NDA). “The use of EDA as part of population pharmacokinetic analysis reduces the time spent in performing more detailed analysis. It provides the analyst the information needed for ‘a straight to the point analysis,’ so they don’t have to test all of the possible relationships between pharmacokinetic parameters and possible predictor variables (covariates). S-PLUS has proven to be a good resource for prototyping pharmacokinetic experimental designs,” says Dr. Ette.

For example, when Dr. Ette explored pharmacokinetic design in a recent study he pooled data from 138 subjects in six studies involving both sexes, approximately 16 percent of which were renal impaired. They ranged in age from pediatrics to the elderly. These subjects were treated with a test drug in different studies and the data was used to develop a model characterizing the disposition of the drug (i.e., time course of drug) in the patient population. “S-PLUS provided us with the tools (data visualization and modern regression techniques) we needed for the initial exploratory examination and analyses of the data to discover patterns and relationships that set the stage for the analyses procedure,” says Dr. Ette.

Further, an exploratory examination of the raw drug concentration data revealed the probable presence of two subpopulations (see figure 1). Dr. Ette then fit a model to the data to characterize the dose-concentration relationship – in the population of subjects studied. With this characterization, individual parameter estimates were generated. Dr. Ette explored his data using modeling techniques in S-PLUS to determine possible covariates that could be used to explain variability in dose-concentration relationship observed in the patients.

Researchers analyzed the distribution of the rate at which the drug was cleared from the body per unit time (termed clearance) in the study population using a probability density plot. [Drug clearance determines drug levels on multiple dosing.] This revealed the existence of two subpopulations (figure 2) as was suggested by the concentration-time profile plot. Multiple linear regression models, generalized additive models [(GAM), a modern regression technique], and tree-based models implemented in S-PLUS were very useful for selecting covariables that could be used to explain the variability in the dose-concentration relationship. [Generalized additive models are a group of models that are as tractable as the linear model but do not force the data into unnatural scales. Separate functions are introduced to allow for nonlinearity and heterogeneous variances. GAM is closer to a reparameterization of the model than a re-expression of the response.] The existence of two subpopulations and covariates that could be used to explain this variability in dose-concentration relationship was captured with these exploratory modeling (data fitting) techniques.

The inadequacy of the linear model is clearly seen in figure 3. The index of renal function (creatinine clearance) was found to be the major determinant of the variability in the dose-concentration relationship and the covariate explaining the subpopulation differences. The contribution of age was judged to be of minimal importance. The dendrogram from the tree based model with the associated histogram drove home this point (see figure 4). The tree-based model confirmed the inadequacy of trying to use linear models to explain the relationship between creatinine clearance (CLCR) and clearance (CL) and confirmed the uncovered structure in the data (i.e., the existence of two subpopulations). “The whole population analysis was done in a much shorter time than is traditionally the case,” Dr. Ette remarks.

Visualization of Data Analysis Results

Data analysis software such as S-PLUS, with its data visualization capability, provides scientists with a powerful tool for communicating results. Failure to deliver these findings successfully can endanger all data analyses efforts, despite the quality of the work. The efficient and timely delivery of information is critical to efficient drug development. S-PLUS conditioning plots and Trellis graphics can be very useful to scientists in communicating their results effectively to their associates. Conditioning plots can be very useful for visualizing and presenting the results of a clinical trial simulation study in which various study designs are investigated to determine the design most likely to produce the desired outcome. “S-PLUS offers interoperability with leading business reporting packages such as Microsoft Office, Excel and Power Point, providing the perfect platform for the presentation of results,” says Dr. Ette.

Next Steps

Efficiency in drug development can be achieved by understanding the sources of variability and accounting through better study design, improved data analysis and better study monitoring. S-PLUS offers researchers a powerful tool for discovering knowledge hidden in study data and understanding the consequences of different study designs. This next-generation software offers scientists the flexibility and power to analyze data faster for early decision making, thereby saving time and money. The use of this software would aid faster drug development, making it possible for safer and more effective drugs to be brought to the market sooner, rather than later.

Merck Deploys Web-Based Visualization and Data Analysis Solution Enhancing Drug Discovery Efforts

Business Challenge

Merck Research Laboratories (MRL), in Rahway, New Jersey, is the foundation of discovery and development on which Merck & Co., Inc. (NYSE:MRK) continues to build innovative drugs to treat a wide variety of diseases including high blood pressure, osteoporosis, asthma, HIV/AIDS, and osteoarthritis. The company also makes vaccines ranging from chicken pox to hepatitis.

MRL is recognized for its contribution as a global provider of beneficial pharmaceutical products developed in its laboratories and for the wealth of knowledge that its scientists has uncovered and shared throughout the scientific and technical world. Creative spirit, dedicated effort, and application of new technology form the basis for innovation that the pharmaceutical industry requires for success in today’s biomedical research environment.

Discovering and producing drugs and vaccines across such a wide spectrum of diseases is expensive and requires major research efforts to keep up with revolutionary developments in modern biology and medicine. To this end, Merck maintains several state-of-the-art research facilities in North America, Europe, and Japan.

Industry analysts estimate that it costs on average $500 million to develop a new drug. Further, the drug development process may take an average of ten years or more. Technology that can accelerate identification of promising new drugs is providing leading pharmaceutical companies with a significant competitive advantage.

As in many other disciplines, advances in automation and miniaturization have enabled researchers at these labs to vastly increase the rate and scale of their work. New technology, such as High Throughput Screening (HTS), is allowing researchers to test hundreds of thousands of samples from Merck’s chemical and natural product “libraries” to determine how they affect disease processes. Traditionally, researchers tested a few hundred samples per day. With technological advances, researchers are now able to test tens of thousands of samples per day, creating difficult challenges for monitoring and maintaining the quality and efficiency of these highly automated, high volume HTS processes.

Dr. Bill Pikounis, Associate Director, and his colleagues at MRL are responsible for building and deploying data analysis solutions using StatServer® technology that helps scientists with limited statistical knowledge produce a short list of promising chemical samples or entities. The software tools are constructed by working closely with HTS scientists. An important goal is to drive down rates of “false positives” — avoiding apparent potency in entities when there is none — and “false negatives” — missing really potent compounds due to process variability.

Business Solution

“We were interested in a solution that could provide powerful analytical tools and deliver the solution to the desktops of our leading scientists. Further, the technology needed to be useful, scalable, easy-to-use, and reliable,” says Pikounis. “We selected StatServer technology because it offered the analytical and visualization tools we needed to develop a solution that could be deployed enterprise wide using a familiar standard Web browser.” StatServer’s Web-based architecture allowed the team to make changes or additions to the scientists’ analytical toolbox quickly in response to their scientific customer’s needs.

StatServer provides a reliable, effective, intelligent solution for accessing, analyzing, and visualizing information for entity selection. Merck statisticians use S-PLUS® from their desktops to develop the data analysis and statistical functionality necessary to evaluate data from new chemical samples. Then, the code is programmed and connected to a familiar Web browser where scientists can access reliable, powerful tools, allowing them to do valid, sensitive, and productive data analyses themselves. “We had to find a way to deploy the solution to our HTS researchers throughout the world,” Pikounis says. “Putting software on individual desktops wasn’t really an option, because we anticipate continuing development and improvement to keep up with the rapidly changing technology. Maintaining an installed base of software over our geographically dispersed user base would have been impossible. That’s why we turned to StatServer.”

By working closely with HTS scientists, the Merck team was able to define critical data-driven issues and, working within the S-PLUS environment, prototype solutions easily, test them on real data, and then demonstrate the technology to scientists for feedback. The Merck team embarked on an ongoing cycle of development and customer feedback to refine the technology solution.
S-PLUS’s extensive built-in data analysis and graphics capabilities, as well as its easy-to-program object-oriented S language, allowed researchers to make changes and incorporate new ideas easily. “In fact, I doubt that we could have done it without S-PLUS,” says Pikounis.

“StatServer has become a cornerstone part of our department operations and has revolutionized how we work with our scientific colleagues,” says Pikounis. Traditionally, the pre-clinical research environment in pharmaceutical companies has only permitted one-to-one type consulting models between statisticians and scientists. “Our HTS data analysis solution using StatServer technology has allowed us to leverage our services by providing our scientists with the tools they need to create a ’short list’ of prospective chemical entities. We are empowering our colleagues and using our resources more effectively.

Communicating research results is essential for effective drug discovery and development. “We selected StatServer technology because it offers superior data graph capabilities. Informative and relevant visualization tools ensure effective study, interpretation, and communication of results amongst colleagues,” says Pikounis.

The technology can be deployed enterprise wide, so researchers can access the information on a global basis. “The flexible, powerful data analysis solution allows us to handle requests from Merck researchers internationally from Spain to San Diego (U.S.A.) across nine different time zones,” says Pikounis. This foundation has provided more challenges and opportunities. Evolving HTS tools and exploding data analysis needs in areas such as genomics and pharmacology have led to other areas where Merck’s solution can be leveraged. The success of StatServer for HTS has provided a model to meet these needs.

“We think StatServer really changes the way we work,” Pikounis commented. “Revolutionary technology has brought large and/or complex data sets to many more areas of drug discovery, research, and development. There is just no way that a small group of statisticians can adequately serve all these needs by working exclusively in a traditional one-to-one consulting mode. Focusing on critical areas and working with our colleagues to develop and deploy analytical tools to scientists with S-PLUS and StatServer allows us to leverage our effectiveness and expand our impact.”

Benefits

Centralized data and analysis capabilities
Improved productivity
Desktop delivery in a familiar environment
Seamless integration with Microsoft Office and other business tools
Business Tools

StatServer
S-PLUS

S-Plus Software Mines Genomic Data

Unix Reviewer: Todd Wood, Director of Bioinformatics, Genomics Institute

Background: Cereals are the most important food crops in the world and determining the entire genomic sequence of a model cereal, such as rice is critical to meeting our future nutritional demands and food security needs. Rice is the single most important food crop in the world feeding over half of the world’s population. The Genomics Institute uses sophisticated data mining techniques to using S-PLUS software to mine genomic data generated by rice.

Problem Solved: Researchers at the Genomics Institute were interested in studying genomic patterns in rice. Discoveries are made by mining genomic data using mapping and sequencing techniques. Todd Wood was interested in using powerful data analysis software to query his genomic data to make important discoveries. He selected S-PLUS 5.1 for Unix because it provided him with powerful analytical tools and unique Trellis graphics. “I selected S-PLUS for UNIX 5.1 because the software is based on the powerful next generation object-oriented language from Lucent Technologies. The S language has always been regarded as the premier language for data analysis and statistical modeling,” says Wood. “With S-PLUS we benefit from superior memory resourcing allowing us to process larger data sets faster. We can pre-process our data and analyze gigabytes of data with modest computer resources.”

Product Functionality: “The product is an invaluable tool for accessing, analyzing and visualizing data. S-PLUS supports sequential processing through block reads and writes, allowing us to analyze arbitrarily large data sets. We have the tolls to handle big problems, from megabytes to gigabytes.

Strengths: The product makes it easy to read data from virtually any source. The comprehensive import/export capabilities reduce time spent moving data from source to source, allowing us to focus on our analysis. S-PLUS also offers a comprehensive set of traditional and modern methods. The software package also offers new statistical techniques including bootstrap, jackknife, robust MM regression, and missing data methods for variance and correlations, giving us industry leading tools.

Selection Criteria: I selected this product because the company has a history of developing and delivering powerful analytical tools. I believe that this product has a significant advantage over other competitors due to its powerful analytics and visualization tools.

Deliverables: The product allows us to query genomic data and to discover significant patterns. The unique Trellis graphics provide easy-to-read reports for comminicating results to colleagues and industry representatives. Further, the graphics allow us to study our data effectively.

Vendor Support: The vendor support has been excellent.

Documentation: Yes, the documentation is complete and easy-to-read.

Environmental Justice Served by Powerful S-Plus and S+Spatial Data Analysis Tool Research

Business Challenge

Environmental justice seeks to ensure that no population is forced to shoulder a disproportionate burden of the negative human health and environmental impacts of pollution or other environmental hazards.

The movement emerged in the early 1980’s in response to large demonstrations opposing the location of a PCB-landfill site in a predominantly black community in Warren County, North Carolina. Continued research and public attention raised concerns of the fairness and protection afforded under existing environmental programs. Today, the White House and EPA considers environmental justice to be a high priority to the extent that the Environmental Protection Agency has created an Office of Environmental Justice. Popular culture reflects the public’s growing interest in these issues with movies such as “Civil Action.”

Frank M. Howell, professor of sociology at Mississippi State University, and John K. Thomas, professor of rural sociology, at Texas A&M University have addressed real-world environmental justice issues with a practical solution by developing an analytical methodology termed “socio-environmental visualization” (SEV). From courtrooms to research facilities, the SEV model has the potential to bring modern scientific statistical analysis to researchers, policy-makers and business decision-makers internationally.

The challenge facing researchers, the legal community, and civic action groups is that inductive procedures alone, such as those provided by maps through GIS, lend themselves to subjective interpretations. “That’s where spatial statistics comes in, putting numbers back on the map,” says Howell.

SEV methodology also relies on scientific data visualization allowing the researcher to better understand the functional forms of the relationships among variables used in the investigations of environmental inequity. A practical problem is to be able to field the SEV methodology on a common computer platform with near real-time execution so that questions can be answered as research takes place, rather than wait for data-translation from one program to another to be conducted.

Business Solution

The SEV model relies on scientific data visualization and comprehensive data analysis of environmental and sociological variables. The researchers are utilizing EPA environmental monitoring data along with Census Bureau information to create a “proof of performance” for their environmental justice paradigm. After two years of preliminary research, the methodology appears to be a powerful evaluation tool providing analyses with greater statistical confidence than before. The SEV methodology relies on three software packages – ArcView® for GIS, S-PLUS® and S+SpatialStats® – to implement the tools of GIS, scientific data analysis, and spatial statistics into an integrated “real-time” methodology, all on one computing platform.

“We selected S-PLUS because it was the package of choice for modern statisticians that includes the most current statistical models,” says Howell. “S-PLUS allowed us to stay within a familiar environment when moving ArcView GIS tabular data back-and-forth for analysis.” Moreover, the ability to perform spatial statistical analyses on the data from an ArcView coverage provided the researchers with productivity gains over other software packages which are not integrated into one platform. The software also provided the researchers with a powerful data analysis tool that could provide publication-quality graphics.

“S+SpatialStats provided several benefits to our research. We could read data in and out of the packages without having to import or export data. The module included the most modern spatial statistical models available. And, results could be viewed in ArcView where new variables could be added to enhance the model’s productivity,” says Howell. Designed specifically for the exploration and modeling of spatially correlated data, S+SpatialStats allowed researchers to easily access leading statistical models for analyzing geostatistical data, lattice data and spatial point patterns. SEV methodology promises to bring greater scientific credibility to the environmental justice work performed by researchers, the legal community and civic action groups.

Benefits

Spatial Correlation and Regression
Point Pattern Analysis
Exploratory Data Analysis
Predictive Modeling
Research Application
Environmental Research
Geography
Business Tools

S-PLUS for WINDOWS
S+SpatialStats
ArcView For GIS

More Areas of Expertise

Insightful offers data analysis consulting services from statistical analysis and model building, to custom application development, to end-to-end system integration. Drawing on its technical consulting and research staff of software engineers, statisticians, and scientists (including 50 Ph.d.’s), Insightful has the expertise to tackle large scale and challenging problems involving data and statistical analysis.

While our group’s core competence is high-end analytics, we have a track record of successful enterprise deployment using the Web to enable frictionless access to information.

Statistical Consulting Expertise:

Analytic Customer Relationship Management
Financial/Business Performance Management
Operations/Production Management
Reporting and Decision Support Systems

Manufacturing
Statistical process control
Survival and reliability analysis
Degradation analysis
Designed experiments and ANOVA
Six Sigma methodology

Finance
Time series analysis
GARCH modeling and volatility forecasting
Robust and nonparametric financial modeling
Time series database management
Environmental statistics
Portfolio optimization

Pharmaceuticals and Health
Survival analysis
Clinical trial design and analysis
Pharmacokinetics
Spatial statistics
Sequential clinical trial analysis
Missing data

Science and Engineering
Signal and image processing
Wavelet analysis
Information retrieval

Image Processing
Search engines

Data Mining and Market Research
Tree modeling
Cluster analysis
Multidimensional scaling
Analysis with Trellis™ graphics
Neural networks
Response modeling

Software Engineering Expertise
Build custom S-PLUS libraries using S-PLUS functions with C, C++, and FORTRAN libraries and graphical user interfaces using the S-PLUS menu and dialog building tools
Develop point-and-click interfaces using JAVA, Visual Basic, Visual C++ and Web programming
Implement client-server software applications using
S-PLUS support for COM/DECOM/ActiveX, S-PLUS DLL’s

S-PLUS Server
Integrate databases with S-PLUS using ODBC, JDBC, S+SDK, or the Informix Universal Server S+DataBlad

Areas of Expertise

Our consulting group offers a wide range of services including statistical analysis and model building, custom S-PLUS and Insightful Miner software development and end-to-end system integration.

Insightful Consulting

Insightful’s consulting services have extensive experience building industry-leading analytic solutions to solve today’s most challenging business problems. From credit risk scoring solutions to regulatory compliance Insightful’s solutions are proven and deliver high ROI.

Whether you have short-term need to address a mission-critical business problem or an initiative to deploy an analytic solution, Insightful consultants can cost-effectively meet your needs quickly.

Our consultants have backgrounds in mathematics, statistics, data mining software engineering, databases and data warehouses with specific expertise in the S-PLUS product line and client/server enterprise-wide deployment strategies on Windows/NT, UNIX and Linux platforms.

Get up to speed quickly! Take one of our S-PLUS Courses

S-PLUS Essentials: The Graphical User Interface is a hands-on training course designed for those new to S-PLUS. This course will provide the fundamentals for data exploration through graphing as well as basic statistics and statistical models.
S-PLUS Essentials: The Command Line, is a hands-on training course designed for users new to S-PLUS. This course provides a full range of command line operations for data manipulation, graphing, statistics and statistical models, and function writing in the S-PLUS language.

Featured Courses

C210: S-PLUS Working with Big Data is a hands-on training course for users who are familiar with S-PLUS scripting including manipulation and summarization of S-PLUS objects, creation of graphics and computing basic statistical summaries using the command line. This course provides thorough instruction in applying various data-analytic techniques to big out-of-memory data sets.