STATools is the Client-Desktop App provided by the GPRO suite to manage a collection of scripts developed in R and other programming languages for statistical analysis and graphical representation of biological data and results (including Tests, ANOVA, Plots, Histograms, Heatmaps, Venn diagrams, Cluster and Diversity analysis, KRONAs, Networks and more). Scripts and all other backend dependencies can be installed in the user PC if it has RAM enough or more properly in a remote server. The software is coupled with the server infrastructure providing the user with a interface-based solution to easily manage the scripts. The App provides a menu to access the interfaces and a left side space to visualize the results.
1.2 - VERSION
The current version of STATools is 2.0, which is distributed within the GPRO suite as an installer for Windows 7 (64 bit), a self-extracting disk image for Mac OS X 10.6 or later (64 bit), and a compressed tarball archive for Linux 2.6 kernel series or later (64 bit).
1.3 - INSTALLING STATOOLS IN YOUR PC
STATools is Java standalone software that can be installed in personal computers (PCs) with 2GB of RAM or higher as well as installation of a Java JRE (Java Runtime
Environment) version 6 or superior.
To check if you already have a JRE installed on your computer:
Open a command line interface.
Type: java -version
Then, you should see a message like the following:
$ java -version
java version "1.8.0_92"
Java(TM) SE Runtime Environment (build 1.8.0_92-b14)
Java HotSpot(TM) 64-Bit Server VM (build 25.92-b14, mixed mode)
If you get a “Command not found” error message, it means that JRE is not installed.
For installing JRE, go to the official JRE repository here and download the version
matching your operating system. Once installed, check again the output of java -version command on a command line interface. Sometimes, although the JRE is installed,
it is not set at the path.
To install the Windows version
Extract the archive using archiver utility program e.g WinRAR. Then browse to the executable file “STATools.exe” and execute/run it.
To install the Mac version
Extract the archive to the desired destination using:
- tar xvf statools.x.y.os.zip
Then browse to the executable binary file “STATools.app” and execute/run it.
To install the Linux version
Extract the archive to the desired destination using:
- tar xvf STATools.x.y.os.zip
Then browse to the executable binary file “STATools” and execute/run it.
1.4 - GETTING FAMILIAR
The layout of STATools is based on the following features; “Directory Browser”, “Working Space”, “Results Preview” and “Menu”.
DIRECTORY BROWSER: To set any folder of your PC as directory for storing the files managed by STATools.
WORKING SPACE FOR INTERFACES: Central working space to launch the interfaces of all R scripts.
MENU: To choose the script and analysis.
[Directory → Select directory folder]: select the Directory Browser.
[Directory → Show]: view the Directory Browser.
[Directory → Hide]: hide the Directory Browser.
[Graphical Analyses → Heatmap]: to create heatmap representations with or without clustering analysis
[Graphical Analyses → Cluster Analysis]: to perform cluster analyses using different, methods and correlation algorithms
[Graphical Analyses → PCA]: to perform Principal Component Analyses at the individual or sample level
[Graphical Analyses → Venn Diagram]: To create Venn diagrams
[Plots → Time series plot]: to create time series plots with multiple variables
[Plots → Scatter plot]: to create scatter plots between two variables (X and Y).
[Plots → Barplot]: to create vertical or horizontal barplots
[Plots → Grouped Barplot]: to create grouped barplots of rows considering columns as grouping references
[Plots → BoxPlot]: to perform and represent a box plot analysis.
[Tests → Chi Square]: to perform chi square tests either as a Test of Independence or as a goodness of fit test.
[Tests → Fisher Exact Test]: to perform fisher exact tests for analisis of contingency tables.
[Tests → ANOVA]: To perform ANOVA tests
[Tests → SNK only NOVA]: to perform a Student Newman-keuls (SNK) test to identify sample means that are significantly different from each other.
[Tests → Tukey only NOVA]: to perform a Tukey test (alternative to SNK) to find sample means that are identify sample means that are significantly different from each other.
[Tests → Multiple Testing Correction and FDR]: to re-calculate probabilities from a statistical test which was repeated multiple times to retain a prescribed family-wise error rate α in an analysis involving more than one comparison. Four methods are provided: Bonferroni correction, Benjamini Hotchberg False Discovery Rate (FDR), Adaptative Benjamini Hotchberg and Storey Qvalue.
[Microbiota Specific → Diversity Indices]: to infer diversity indices from 16s/18s amplicons using different algorithms
[Microbiota Specific → KRONA]: to create KRONA graphical representations of relative abundances and confidences within the complex hierarchies of metagenomic classifications
[Microbiota Specific → Unifrac]: to compute differences between microbial communities based on phylogenetic information using Unifrac.
STATISTICS & METRICS:
[Statistics & Metrics → Summaries]: To infer a summary of statistics from a set of counts
[Preferences → Proxy Connection Settings]: setup proxy configuration for server connection
[Preferences → Pipeline Connections Settings]: to setup your user credentials to access the server
[Help → Manual]: link to this manual.
[Help → About STATools]: Other technical details and copyrigth of STATools.
1.5 - SERVER SIDE DEPENDENCIES
STATools is a Client Side + Server Side solution thus meaning that the application is coupled via API with a bioinformatic infrastructure called GPRO Server Side that contains all the dependencies needed by VariantSeq to execute the workflows and pipelines. These dependencies are scripts, databases and the following third party CLI software:
Linux Operating System with minimum Bash version 4.
Apache HTTP server 2.2 or later.
R (Recommended version 3.4)
Other scripts created by Biotechvana to handle clients connections and run jobs/pipelines.
The GPRO Server Side can be installed in the PC of the user or in remote servers as a Cloud Computing resource. However, its installation is a complex task due to the lot of dependencies and requirements (besides of the CLI software) for installing and running this infrastructure. For this reason, we distribute the GPRO Server Side in a Docker container that can be easily installed for the user in a couple of steps. Indications for installation of the GPRO Server Side Docker are available here.
1.6 - LINKING STATOOLS WITH THE SERVER SIDE
Once the GPRO server side docker has been installed you need to link VarianSeq to it. To do this, go to [Preferences → Pipeline connection settings] in the top menu and type the following into the configuration Dialog (Fig.2):
Your email address: to receive notifications from the server.
Host address: here you should type localhost (see figure 2).
Port number: This field should be filled inly in case of you run the server side manually. The default number will be 22.
User name and password: Your ID credentials provided to access the host server.
As also shown in Figure 2 you can also check the option “Run GPRO server locally using Docker” to let you to automatically start the GPRO container each time you run VariantSeq (Also note that if you have this option checked you do not need to type the port). You can test if the app is connected to the Server Side clicking on the tab “Test connection settings”. Alternatively, if you install the Server Side manually (without the Docker) just add the IP of the remote server where the Server Side is hosted, add the port information (by default 22) and keep the Option “Run GPRO server locally using Docker” unchecked
In case you would need to connect the server infrastructure via Proxy, STATools will require to be configured. You can choose one of the three following methods:
If you do not know the proxy settings, choose the "Use system proxy settings" to let STATools to guess the default proxy settings already configured in your computer.
If you know your proxy settings, you can specify the proxy configuration. This is the preferred option when using a network proxy. User, password and FTP settings are optional. The port for HTTP proxy is usually 8080 by default.
If you have a Proxy Automatic Configuration file (.pac) URL, use it for loading settings automatically from a remote file.
1.7 - RAM ASSIGNATION TO YOUR PC
The RAM assigned to STATools can be modified editing two parameters (Xms and Xmx) in the configuration file named “STATools.ini”. In Linux and Windows computers, the “STATools.ini” configuration file is located inside STATools app folder. On macOS computers, this configuration file can be found by right-clicking on STATools.app → Show package contents → Contents → MacOS → STATools.ini.
Within the “STATools.ini” file, Xms and Xmx parameters look like this;
The values correspond to the assigned RAM in Megabytes. To modify the RAM assigned to STATools, just change these values. Please keep in mind that the limitation depends on the amount of RAM of your computer. E.g. if your computer has 8GB of RAM, it is recommended to assign Xms2048m and Xmx4096m for a better performance. Xmx can be even increased up to Xmx6144m.but please avoid to use RAM values near to the maximum available memory of your PC as this election might provoke unstability in computer’s operating system.