Open-source ecosystem for in vitro and microphysiological systems
Save the DateDecember 1-2, 2026Charlotte, North Carolina12th Annual Regenerative Manufacturing Innovation Consortium (RegMIC) Meeting: A Roadmap for Tissue Organoids and Body-on-a-Chip Manufacturing Standards
Back to Home

Transcriptomics tool

Transcriptomics Biomarker Identification

Use the GLMQL-MAS workflow to run differential expression, rank biomarker candidates, select top genes per comparison, and compare PCA results from all normalized genes versus top MAS-selected genes.

Transcriptomics biomarker identification

Run the original GLMQL-MAS workflow online

This tool helps users find genes that change meaningfully between a baseline group and one or more comparison groups. It starts from a bulk RNA-seq count matrix, identifies differentially expressed genes, ranks candidate biomarkers with GLMQL-MAS, and prepares tables, PCA plots, pathway summaries, and downloadable result files.

In plain language, GLMQL-MAS helps narrow thousands of genes down to a focused list of candidates that show both strong expression change and strong statistical evidence. Users can run the built-in example dataset or upload their own CSV file, then adjust the same settings used in the original workflow.

Default analysis settingsM = 1, A = 1, alpha = 0.05, FDR cutoff = 0.05, |logFC| cutoff = 1, and top genes = 10.

Required CSV format before you start

Prepare your RNA-Seq count matrix correctly

To use this software, you need to have the RNA-Seq count data after alignment ready in a CSV file where the rows are genes and the columns are samples.

Column structure

The first column should be titled Gene Symbol and should contain gene names. The next columns should represent samples from different groups.

Sample naming rule

The software recognizes samples within a group by the unique name before the parentheses. The number inside parentheses indicates the sample number.

Example with three groups
  • Group A has 10 samples.
  • Group B has 15 samples.
  • Group C has 10 samples.

The sample columns should be named as: A (1), A (2), ..., A (10), B (1), ..., B (15), C (1), ..., C (10).

Gene SymbolA (1)A (2)B (1)C (1)
GeneA12013428095
GeneB908830115
Please make sure your CSV file is formatted this way before uploading it. If sample names do not follow the group-name-plus-parentheses pattern, group detection may be incorrect.

Question 1

How many treated groups does your experiment have?

This follows the original software choice between one treated group and more than one treated group.

Expected downloadable files

TMM_normalized_counts.csv
DE_<baseline>_vs_<treated>.csv for each comparison
Significant_DE files after alpha, FDR, and |logFC| filtering
Ranked_DE files with MAS_Score and MAS_rank
Top gene files for each selected treated group
Common and unique top-gene table across treated groups
PCA outputs for all genes and top MAS-selected genes
Optional GO and Reactome pathway analysis files

Standards and manufacturing partners

MOSSDO and RegMIC

Connect with the organizations advancing consensus standards, validation, and regenerative manufacturing across the PhysioVerse ecosystem.