Logical Clustering Suite v0.5beta (Win PCs)

[niethamp]

### Simple Gene Correlation Analysis (sGCA) by Permutated Logical Clustering

This is a standalone, beta version of the Logical Clustering Suite (v0.5) for Win PCs. Its installation requires download of MATLAB runtime as wrapper.

• The Logical Clustering Suite (LCS) clusters gene expression profiles or similar data by permutated logical gating according to their “Ideal Phenotypes” (IPs), which are defined by all possible experimental outcomes.
• Logical clustering conceptually differs from K-means-, SOM, DBSCAN and alike clustering methods that cluster gene expression profiles just according to their mutual similarity without taking the experimental groups into account.
• When just comparing two experimental groups, logical clustering simplifies to something like DESeq2 with only two possible IPs, 0 1 for upregulation & 1 0 for downregulation. Thus, methods like DESeq2, may be conceptualized as a special instance of logical clustering.
• In summary, logical clustering assumes that the locations & number of all experimentally meaningful cluster centers are given by the experimental design. Gene expression profiles more similar to one IP than to all the other IPs, form a logical cluster.
• Logical clustering by simple (=logic) gene correlation analysis (sGCA) was introduced in Ma Y, Hui KL, Gelashvili Z, Niethammer P. Oxoeicosanoid signaling mediates early antimicrobial defense in zebrafish. Cell Rep. 2023 Jan 31;42(1):111974. doi: 10.1016/j.celrep.2022.111974. Epub 2023 Jan 10. PMID: 36640321; PMCID: PMC9973399. Please cite if you are using LCS.

To help improving this applications, please send feedback to sgcafeedback@gmail.com.
If you like LCS, please follow us on https://twitter.com/NiethammerLab

---

This discussion was created from the release Logical Clustering Suite v0.5beta (Win PCs).