Oprogramowanie

Tackle

Tackle is a Shiny Server based visualization dashboard for analyzing Broadside results. Tackle allows to manipulate interactive feature interaction networks, perform literature search and to perform ad-hoc GO and KEGG overrepresentation tests. Pre-release.
[Author: Łukasz Król]

Broadside

Broadside is a distributed feature selection and interaction mining software for classification, regression and survival analysis. It can work on high-dimensional problems of millions of features as well as on smaller sets of well chosen clinical or PCR variables. Both numerical and categorical variables are supported. Broadside can be ran on desktop as well as on a clusters containing hundreds of processors.
[Author: Łukasz Król]

MiMSeg

The package contains algorithm for automated detection of tumor tissue on NMR apparent diffusion coefficient maps. The algorithm itself may be sued for segmentation of any lesions on basically any medical image, however adjustments are required.
After sucessful usage please cite:

Binczyk Franciszek, Bram Stjelties, Christian Weber, Michael Goetz, Klaus Maier-Hein, Hans-Peter Meinzer, Barbara Bobek-Billewicz, Rafal Tarnawski, Joanna Polanska: MiMSeg - an algorithm for automated detection of tumor tissue on NMR apparent diffusion coefficient maps. Information Sciences, 2017, 384:235-248, doi:10.1016/j.ins.2016.07.052
[Author: F. Binczyk, J.Polanska]

Spec-GMM

Matlab implementation of the efficient algorithm for Gaussian mixture modeling of spectra of different types (e.g., MALDI-ToF profiling, MALDI-IMS, NMR spectroscopy). The main idea is automated partitioning of spectral signal into fragments. The obtained fragments are separately decomposed into Gaussian mixture models. The parameters of the mixture models of fragments are then aggregated to form the mixture model of the whole spectrum. https://github.com/ZAEDPolSl/Spec-GMM
[Author: Polanski, Marczyk, Polanska]

2DGMMgel

A three step algorithm for analysis of 2D gel electrophoresis images. The main idea is based on improving the efficiency of spot detection performed with any algorithm, by fitting mixture of 2D normal distributions. The algorithm is implemented in MATLAB R2016b. https://github.com/ZAEDPolSl/2DGMMgel
[Author: Michał Marczyk]

MrGSEA

MATLAB implementation of Gene Set Enrichment Analysis (GSEA) method for finding pathways which can show more complex relation between changes of gene expression due to different phenotypes. Major adjustments of the method:
- 12 gene ranking metrics implemented (plus its ranks and/or absolute values)
- possibility to use external ranking metric method (by applying own function)
- possibility to use external Gene Set Database
- phenotype permutation available for all ranking methods (even external ranking)
- increased efficiency by parallel computing used
https://github.com/ZAEDPolSl/MrGSEA
[Author: Michał Marczyk]

BatchI

R package with the batch effect identification by dynamic programming algorithm implementation.
[Author: Anna Papież]

GaMRed

GaMRed is a fully automatic application for filtering insignificant features, which can process different kinds of high-throughput data e.g. microarray, qRT-PCR or RNA-Seq. The method is based on adaptive algorithms for estimation of the cut-off threshold using Gaussian mixture model and k-means clustering. The proposed software supports both Windows and Mac OSX operating systems.
https://github.com/ZAEDPolSl/GaMRed
[Authors: Marczyk M, Jaksik R, Polanski A, Polanska J]

DiviK

Divisive Intelligent K-Means algorithm (DiviK) for joint feature selection and clustering of heavily multidimensional data.
GitHub: https://github.com/gmrukwa/divik
Documentation: https://sut-data-mining.gitbook.io/divik
Docs: https://divik.readthedocs.io/en/latest/index.html
PyPI library: https://pypi.org/project/divik/
Docker image: https://hub.docker.com/r/gmrukwa/divik
[Author: Grzegorz Mrukwa]