Interactive Platform for Statistical Computing - HelikarLab/StatisticalPlatform GitHub Wiki

Interactive Platform for Statistical Computing is a tool designed by Helikar Lab (Computational Biology @ University of Nebraska-Lincoln) to provide a free web-based platform to perform various statistical computations without having any knowledge of the high level programming languages, which is required for most statistical technologies. Most statistical technologies require users to be familiar with the command line and/or some type of a higher level programming language, making statistics less accessible to those who are not familiar with these technologies.

The goal of this project is to develop the final version of a free web-based platform that enables anyone to perform various statistical computations in an easy-to-use, interactive, and graphical manner. During Google Summer of Code (GSoC) 2015, the first version of a web-based platform that utilizes the R server to perform interactive statistical analyses was built. Specifically, the following statistical functions were implemented in the first phase: Tabular data upload/export/visualization, descriptive statistics, t-tests (single-sample, independent sample, paired), graphing (2D/3D scatter plots, line, bar charts, box plots, histograms) and one-/multi-way ANOVA.

As part of the second phase in GSoC 2016, the focus was to implement the statistical functions such as clustering, classification, time series analysis, regression models, correlation & covariance computations, heatmaps and various other plots.

The tool has been deployed and is currently in its beta phase.

Tool: http://129.93.136.33/ocpu/library/StatisticalPlatform/www/components.html

GSoC - 2016

Contributor: Shubham Kumar ( [email protected] )

Mentor: Dr. Akram Mohammed

Following are the salient features that are implemented in this project. Most of the features are successfully done and merged. While there are certain new features which were also introduced during the course of project.

For, the complete list of commit, visit: https://github.com/shubhamkmr47/Helikar/commits/master

Commit Description Merged Status
hierarchical clustering dendogram ✔
kmeans clustering ✔
scatter matrix plot ✔
R integration with JS using OpenCPU ✔
time series analysis (single/multiple) ✔
qqplot with qqline ✔
correlation and covariance matrix ✔
simple, grouped and stacked bar chart ✔
scatterplot with linear, exponential, logarithmic and polynomial trendlines ✔
heatmap ✔
download result as SVG ✔
import data from URL ✘
edit and export data option ✔
box plot interpretation ✔
use .tsv file as input ✘
linear and multiple regression analysis ✔
dashboard ✔
R scripts view and sample inputs ✔

Future Works

Presently, the tool is in its beta version and is ready for deployment. There are certain work which is planned to be done in the future alongside the testing phase.

User Management

  • User registration
  • User profile
  • User login
Analysis
  • density plot
  • distribution-based clustering
  • density-based clustering
  • principal component analysis
  • support vector machines
⚠️ **GitHub.com Fallback** ⚠️