Open-source Statistical Language Research Articles

Analyzing dynamic data: A tutorial

Modern data collection techniques allow for intensive measurement within subjects. Analyzing this type of data requires analyzing data at the within subject as well as between subject level. Although sometimes conclusions will be the same at both levels, it is frequently the case that examining within subject data will show much more complex patterns of results than when they are simply aggregated. This tutorial is a simple introduction to the kind of data analytic strategies that are possible using the open source statistical language, R.

Innovating to enhance clinical data management using non-commercial and open source solutions across a multi-center network supporting inpatient pediatric care and research in Kenya.

Objective To share approaches and innovations adopted to deliver a relatively inexpensive clinical data management (CDM) framework within a low-income setting that aims to deliver quality pediatric data useful for supporting research, strengthening the information culture and informing improvement efforts in local clinical practice. Materials and methods The authors implemented a CDM framework to support a Clinical Information Network (CIN) using Research Electronic Data Capture (REDCap), a noncommercial software solution designed for rapid development and deployment of electronic data capture tools. It was used for collection of standardized data from case records of multiple hospitals’ pediatric wards. R, an open-source statistical language, was used for data quality enhancement, analysis, and report generation for the hospitals. Results In the first year of CIN, the authors have developed innovative solutions to support the implementation of a secure, rapid pediatric data collection system spanning 14 hospital sites with stringent data quality checks. Data have been collated on over 37 000 admission episodes, with considerable improvement in clinical documentation of admissions observed. Using meta-programming techniques in R, coupled with branching logic, randomization, data lookup, and Application Programming Interface (API) features offered by REDCap, CDM tasks were configured and automated to ensure quality data was delivered for clinical improvement and research use. Conclusion A low-cost clinically focused but geographically dispersed quality CDM (Clinical Data Management) in a long-term, multi-site, and real world context can be achieved and sustained and challenges can be overcome through thoughtful design and implementation of open-source tools for handling data and supporting research.

Survival Prediction for Pancreatic Cancer Patients Receiving Gemcitabine Treatment

Although gemcitabine monotherapy is the standard treatment for advanced pancreatic cancer, patient outcome varies significantly, and a considerable number do not benefit adequately. We therefore searched for new biomarkers predictive of overall patient survival. Using LC-MS, we compared the base-line plasma proteome between 29 representative patients with advanced pancreatic cancer who died within 100 days and 31 patients who survived for more than 400 days after receiving at least two cycles of the same gemcitabine monotherapy. Identified biomarker candidates were then challenged in a larger cohort of 304 patients treated with the same protocol using reverse-phase protein microarray. Among a total of 45,277 peptide peaks, we identified 637 peaks whose intensities differed significantly between the two groups (p < 0.001, Welch's t test). Two MS peaks with the highest statistical significance (p = 2.6 x 10(-4) and p = 5.0 x 10(-4)) were revealed to be derived from alpha(1)-antitrypsin and alpha(1)-antichymotrypsin, respectively. The levels of alpha(1)-antitrypsin (p = 8.9 x 10(-8)) and alpha(1)-antichymotrypsin (p = 0.001) were significantly correlated with the overall survival of the 304 patients. We selected alpha(1)-antitrypsin (p = 0.0001), leukocyte count (p = 0.066), alkaline phosphatase (p = 8.3 x 10(-8)), and performance status (p = 0.003) using multivariate Cox regression analysis and constructed a scoring system (nomogram) that was able to identify a group of high risk patients having a short median survival time of 150 days (95% confidence interval, 123-187 days; p = 2.0 x 10(-15), log rank test). The accuracy of this model for prognostication was internally validated and showed good calibration and discrimination with a bootstrap-corrected concordance index of 0.672. In conclusion, an increased level of alpha(1)-antitrypsin is a biomarker that predicts short overall survival of patients with advanced pancreatic cancer receiving gemcitabine monotherapy. Although an external validation study will be necessary, the current model may be useful for identifying patients unsuitable for the standardized therapy.

A Prospective Gene Expression Profiling Study for Prediction of Cytogenetic Response in Early and Late Chronic Phase CML.

Gene expression profiling has proven useful for identification of gene sets predicting response and/or prognosis in various hematological malignancies. So far expression profiling studies in CML predicting response to imatinib therapy suffered from major flaws, as previous retrospective studies relied on various technical platforms using rather small patient populations. Here, we present the interim results of an expression profiling study performed within a prospective multicenter CML trial of the Central European Leukemia Study Group(CELSG) comparing standard (SD) with high-dose (HD) imatinib as second line treatment for chronic phase CML. Blood samples from at total of 102 CML patients with early or late chronic phase disease were collected prior to and six weeks after treatment with either 400 mg/day for 12 months or 800 mg/day for 6 months followed by 6 months of 400 mg imatinib therapy. Whole blood samples preserved by PAX gene technology were further processed and evaluated centrally at the expression profiling core facility at Innsbruck Medical University. For expression profiling samples were subjected to hemoglobin mRNA reduction before target preparation for hybridization to Affymetrix hGU133 Plus 2 genechips detecting ∼47000 transcripts thereby allowing a whole genome profiling approach. For response prediction we used both the linear discriminatory analysis (LDA) and a random Forest (RF) decision tree algorithm provided in Bioconductor software packages for the open source statistical language “R”. A 100- or 10-fold crossvalidation was performed using a robust leave-10-out setting for LDA and RF respectively. The planned interim analysis of clinical data from 76 evaluable patients undergoing 12 months of therapy with daily doses of 400 mg or 800 mg imatinib revealed 50 (66%) patients with and 26 (34%) patients without a major cytogenetic response (<35% Ph+ metaphases). We found, irrespective of imatinib dosing, a predictive gene expression signature for achieving a major cytogenetic response after 12 months of imatinib therapy. This predictive gene set comprises five genes (ENTPD4, F5, SCL22A4, AIM2, TLR5, PXK) achieving a correct prediction rate of 87% within the 6-week-treatment patient subgroup and an overall prediction rate of 76%. In conclusion, this is the first prospective gene expression profiling study in CML indicating a gene expression profile predicting major cytogenetic response, which still represents a robust response parameter for long-term outcome during imatinib therapy of patients with early or late chronic phase CML.

Using the open-source statistical language R to analyze the dichotomous Rasch model

R, an open-source statistical language and data analysis tool, is gaining popularity among psychologists currently teaching statistics. R is especially suitable for teaching advanced topics, such as fitting the dichotomous Rasch model--a topic that involves transforming complicated mathematical formulas into statistical computations. This article describes R's use as a teaching tool and a data analysis software program in the analysis of the Rasch model in item response theory. It also explains thetheory behind, as well as an educator's goals for, fitting the Rasch model with joint maximum likelihood estimation. This article also summarizes the R syntax for parameter estimation and the calculation of fit statistics. The results produced by R is compared with the results obtained from MINISTEP and the output of a conditional logit model. The use of R is encouraged because it is free, supported by a network of peer researchers, and covers both basic and advanced topics in statistics frequently used by psychologists.

NOISE-ROBUST SOFT CLUSTERING OF GENE EXPRESSION TIME-COURSE DATA

Clustering is an important tool in microarray data analysis. This unsupervised learning technique is commonly used to reveal structures hidden in large gene expression data sets. The vast majority of clustering algorithms applied so far produce hard partitions of the data, i.e. each gene is assigned exactly to one cluster. Hard clustering is favourable if clusters are well separated. However, this is generally not the case for microarray time-course data, where gene clusters frequently overlap. Additionally, hard clustering algorithms are often highly sensitive to noise. To overcome the limitations of hard clustering, we applied soft clustering which offers several advantages for researchers. First, it generates accessible internal cluster structures, i.e. it indicates how well corresponding clusters represent genes. This can be used for the more targeted search for regulatory elements. Second, the overall relation between clusters, and thus a global clustering structure, can be defined. Additionally, soft clustering is more noise robust and a priori pre-filtering of genes can be avoided. This prevents the exclusion of biologically relevant genes from the data analysis. Soft clustering was implemented here using the fuzzy c-means algorithm. Procedures to find optimal clustering parameters were developed. A software package for soft clustering has been developed based on the open-source statistical language R. The package called Mfuzz is freely available.