The web application JaqPotQuattro allows building QSAR models and using them for predictions. As an input, JaqPotQuattro uses data available from data.enanomapper.net. After preprocessing, transforming and preparing the datasets, the various modeling algorithms can be applied (see below).
EXPLORE
Application: http://www.jaqpot.org/
API access: http://app.jaqpot.org:8080/jaqpot/swagger/
User Interface: (under development, coming soon)
Feedback & support: https://github.com/KinkyDesign/JaqpotQuattro/issues
Source code: https://github.com/KinkyDesign/JaqpotQuattro
SUMMARY
Status: Beta
Modeling algorithms available: Multivariable linear regression, Lasso, Elastic net, hierarchical clustering, bi-clustering, ID3 decision tree, partial least squares, partial least squares with VIP variable selection, radial basis function neural networks, support vector machines.
Integrated languages (for algorithm): Python, R, Java (Weka)
Algorithms exposed at third-party web services that obey the API can be integrated.
TUTORIAL
See: D4.3 nQSAR Modelling infrastructure - Section 4, Page 31-38 (to be uploaded and linked)
DOCUMENTATION AND REPORTS
https://www.enanomapper.net/wp/4-analysis-modelling
The tool provides the user with a systematic framework for the automated analysis of microscopy images of nanomaterials and the calculation of nanoparticle descriptors.
EXPLORE
Application: http://app.jaqpot.org:8880/imageAnalysis/
Source code: https://github.com/enanomapper/imageAnalysis
TUTORIAL
Image descriptor calculation web tool
DOCUMENTATION AND REPORTS
https://www.enanomapper.net/wp/4-analysis-modelling
D4.2 Descriptor Calculation Algorithms and Methods
RRegrs is a collection of R regression tools based on the R caret package. It is used to find the best regression models for any numerical dataset. The main use of the script is aimed at finding optimal and well validated QSAR models for chemoinformatics and nanotoxicology.
DOWNLOAD
https://github.com/enanomapper/RRegrs
SUMMARY
Status: Release 0.04 (doi:10.5281/zenodo.21946)
Implemented algorithms:
• Linear Multi-regression (LM),
• Generalized Linear Model with Stepwise Feature Selection (GLM)
• Partial Least Squares Regression (PLS)
• Lasso regression
• Elastic Net regression (ENET)
• Support vector machine using radial functions (SVM radial)
• Neural Networks regression (NN)
• Random Forest (RF)
• Random Forest-Recursive Feature Elimination (RF-RFE)
• Support Vector Machines Recursive Feature Elimination (SVM-RFE)
TUTORIAL
RRegrs Package Tutorial
DOCUMENTATION AND REPORTS
https://www.enanomapper.net/wp/4-analysis-modelling
Tsiliki et al., 2015, RRegrs: An R package for Computer-aided Model Selection with Multiple Regression Models
Chipster is an open source user-friendly analysis software for high-throughput data analysis. It offers over 350 bioinformatics tools and it is constantly updated according to the latest state-of-the-art tools and scripts. Users can analyse and visualize data interactively, and share complete analysis sessions and automatic workflows with colleagues. Chipster supports the analysis of over 120 different microarray platforms, including the most common types by Affymetrix, Agilent and Illumina. In addition, Chipster has extensive functions and tools for the analysis of NGS data.
EXPLORE
Chipster’s client software uses Java Web Start to install itself automatically and connects to computing servers to perform the actual analyses.
Chipster has been installed on several servers around the world. If you would like to use Chipster, you can ask the IT people in your institute to set up a Chipster server for you. This is easy because Chipster is packaged as a virtual machine image. If your institute doesn’t have the required computer hardware, Chipster server can be also installed (free of charge) in the EGI infrastructure for European researchers. EGI and ELIXIR are currently planning to set up also a ready-made Chipster server for end users, so in the future you will not need a local admin person to help you. You can also contact one of the following:
· Finnish users: CSC
· Swedish users: UPPMAX
· Dutch users: Dutch TechCentre for Life Sciences
· German users: DKFZ
· International users: EGI
Chipster website: http://chipster.csc.fi/
Contact developers: chipster@csc.fi
TUTORIALS
- How to Use Chipster for Bioinformatics Analysis of Nanomaterial-Based Omics Data
- Youtube
- ELIXIR Finland
- Demo version for testing: pre-made sessions (username and password: guest) (the account does not allow running analysis)
DOCUMENTATION AND REPORTS
Reports: https://www.enanomapper.net/wp/4-analysis-modelling
lazar is a modular framework for read across predictions of chemical toxicities. Within the eNanoMapper project lazar was extended with capabilities to handle nanomaterial data, interfaces to other eNanoMapper services (databases from data.enanomapper.net and ontologies) and a stable and user-friendly graphical interface for nanoparticle read-across predictions.
nano-lazar:
- mirrors eNanoMapper data for read-across models
- creates read-across predictions for nanoparticles
- uses ontologies (eNanoMapper, BioPortal, UniProt) to explain domain specific terms
- uses ontologies and eNanoMapper data as supporting information for read-across predictions
EXPLORE
Application: https://nano-lazar.in-silico.ch/
API access: https://enm.in-silico.ch/swagger/
Feedback & support: https://github.com/eNanoMapper/nano-lazar/issues
Source code UI: https://github.com/eNanoMapper/nano-lazar/
Source code lazar: https://github.com/opentox/lazar
Source code lazar-rest: https://github.com/opentox/lazar-rest
SUMMARY
Status: V 1.1.0
Technology: Web application, Library / Ruby
License / Waiver: GPL3
Modeling algorithms available:
233 classification and regression algorithms from the R caret package
(https://topepo.github.io/caret/available-models.html)
TUTORIALS
Create and use lazar nanoparticle models https://www.enanomapper.net/library/lazar-nanoparticle
How to reproduce validation results of nano-lazar-paper https://www.enanomapper.net/library/nano-lazar-paper
DOCUMENTATION AND REPORTS
Report: D5.7 Final report on User Applications - Section 3.3.6. NANO-LAZAR, Page 13,17-18,19
Presentation: Validation of read across predictions for nanoparticle toxicities https://enanomapper.net/library/nanoparticles-toxicity
Workshop: Read across toxicity predictions with nano-lazar
https://nano-lazar.in-silico.ch/enm-workshop.html
https://www.enanomapper.net/wp/4-analysis-modelling