MolecularFeaST

The Molecular Feature Selection Tool (MolecularFeaST) is a free-to-use, standalone application that streamlines the use of our feature selection algorithm. MolecularFeaST has been used to find biomarkers based on numerous -omics profiles including: RNA-seq, small RNA-seq, RT-qPCR data, NanoString profiling, microarray data, srRNA expression, bisulfite (DNA methylation) sequencing, and protein mass spectroscopy. Contact us to request a copy of the software.

Import dataset

Import -omics profiles from Excel or .csv files with a single click.

Choose comparisons and feature selection algorithms

Choose the classes you wish to discriminate, and which feature selection algorithms you would like to apply. Choose any number of the 14 available algorithms.

View results and desired features

Use the scatterplot window to view the ranked features and select candidate biomarkers. Select manually or by top percentile.

Generate and save clustering analyses

Create t-SNE and heatmap visualizations from your selected features.

Visualize with other built-in tools

System requirements

Specification	Minimum Requirement	Recommended Requirement
Operating System	Windows 7 SP 1 or macOS 10.14	Windows 7 SP 1 and later, or macOS 10.14 and later (admin install)
Disk Space	10 GB	15 GB or more
RAM	4 GB	8 GB or more
CPU	Any Intel or AMD x86-64 processor	Any Intel or AMD x86-64 with four logical cores or more and AVX2 support

Citation

If you use MolecularFeaST, please cite the following:

Gerolami, J.; Wong, J.J.M.; Zhang, R.; Chen, T.; Imtiaz, T.; Smith, M.; Jamaspishvili, T.; Koti, M.; Glasgow, J.I.; Mousavi, P.; Renwick, N.; and Tyryshkin, K. A Computational Approach to Identification of Candidate Biomarkers in High-Dimensional Molecular Data. Diagnostics 2022, 12, 1997.