Data & Analytics

Mass spectrometry raw data quality control using PTXQC, rawTools, or MSstatsQC.

Differential protein abundance testing using MSstats, limma, proDA, and scipy/statsmodels for Python. Multiple testing correction with BH FDR.

Multi-omics query routing and pipeline orchestration across all OmicsClaw domains. Routes natural language queries to the correct analysis skill across spatial transcriptomics, single-cell omics, genomics, proteomics, and metabolomics.

Statistical analysis for metabolomics — PCA, PLS-DA, clustering, and univariate tests.

Metabolomics data normalization, scaling and transformation.

Metabolomics differential analysis using univariate tests (t-test, FDR), multivariate methods (PCA, PLS-DA, OPLS-DA, sPLS-DA), Random Forest, and ROC analysis for biomarker discovery.

Bulk-to-single-cell trajectory interpolation — uses VAE and GNN to bridge bulk RNA-seq with single-cell reference data, generating synthetic single-cell profiles and embedding bulk samples into developmental trajectories.

Survival analysis for bulk RNA-seq — Kaplan-Meier curves, Cox proportional hazards, expression-based patient stratification.

Alternative splicing analysis — PSI quantification, differential splicing event detection from rMATS/SUPPA2 output.

RNA-seq read alignment and quantification statistics — STAR/HISAT2/Salmon log parsing, mapping rate, unique/multi-mapped reads, library strandedness, gene body coverage.

Bulk RNA-seq count matrix quality control — library sizes, gene detection, sample correlation, outlier detection, CPM normalization.

Protein-protein interaction network analysis from DEG lists — STRING API query, graph construction, hub gene identification.

Differential expression analysis via PyDESeq2 with Welch's t-test fallback — volcano plots, MA plots, p-value diagnostics.

WGCNA-style weighted gene co-expression network analysis — module detection, soft thresholding, hub genes.

Batch effect correction for multi-cohort bulk RNA-seq data using ComBat, with PCA-based visualization before and after correction.

--> --- name: bio-spatial-trajectory description: Trajectory inference and pseudotime analysis using DPT, Monocle3, Slingshot, scVelo for RNA velocity, and PAGA for abstracted graph analysis. tool_type: mixed primary_tool: singlecell measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes. allowed-tools: - read_file - run_shell_command --- You are **SC Trajectory**, a specialised OmicsClaw agent for trajectory inference and pseudotime ordering in single-cel...

--> --- name: bio-sc-preprocessing description: Single-cell RNA-seq QC, normalization, HVG selection, PCA, UMAP, and Leiden clustering. Supports both Scanpy (Python) and Seurat (R) workflows. tool_type: mixed primary_tool: singlecell measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes. allowed-tools: - read_file - run_shell_command --- You are **SC Preprocessing**, the foundation skill for single-cell analysis in OmicsClaw. Your role is to load scRNA-se...

--> --- name: bio-sc-grn description: "Gene regulatory network inference using pySCENIC three-step pipeline\ \ (GRNBoost2 \u2192 cisTarget \u2192 AUCell), with correlation-based fallback. Identifies\ \ transcription factor regulons and scores their activity per cell." tool_type: mixed primary_tool: singlecell measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes. allowed-tools: - read_file - run_shell_command --- You are **SC GRN**, a specialised OmicsCla...

--> --- name: bio-spatial-de description: "Differential expression analysis for single-cell data \u2014 marker\ \ gene discovery using Wilcoxon, t-test, MAST, or DESeq2 pseudo-bulk analysis." tool_type: mixed primary_tool: singlecell measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes. allowed-tools: - read_file - run_shell_command --- You are **SC DE**, the differential expression and marker gene discovery skill for single-cell data. Your role is to id...

--> --- name: bio-sc-communication description: Cell-cell communication analysis via ligand-receptor interaction scoring using CellChat (R), NicheNet (R), LIANA (Python), or built-in L-R database. tool_type: mixed primary_tool: singlecell measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes. allowed-tools: - read_file - run_shell_command --- You are **SC Communication**, a specialised OmicsClaw agent for cell-cell communication analysis via ligand-recept...

--> --- name: bio-differential-expression-batch-correction description: Remove batch effects from RNA-seq data using ComBat, ComBat-Seq, limma removeBatchEffect, and SVA for unknown batch variables. Use when correcting batch effects in expression data. tool_type: r primary_tool: sva measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes. allowed-tools: - read_file - run_shell_command ---

--> --- name: bio-bulkrna-splicing description: "Alternative splicing analysis \u2014 PSI quantification, differential\ \ splicing event detection from rMATS/SUPPA2 output." tool_type: mixed primary_tool: bulkrna measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes. allowed-tools: - read_file - run_shell_command --- Alternative splicing quantification and differential splicing event detection. Accepts pre-computed splicing event tables (e.g. from rMATS o...

--> --- name: bio-bulkrna-de description: Bulk RNA-seq differential expression analysis using PyDESeq2 with optional edgeR/limma-voom via rpy2. tool_type: mixed primary_tool: bulkrna measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes. allowed-tools: - read_file - run_shell_command --- Differential expression analysis for bulk RNA-seq count data. Primary engine is PyDESeq2 (a pure-Python re-implementation of DESeq2); falls back to a scipy Welch's t-test...

--> --- name: bio-bulkrna-coexpression description: "WGCNA-style weighted gene co-expression network analysis \u2014 module\ \ detection, soft thresholding, hub genes." tool_type: mixed primary_tool: bulkrna measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes. allowed-tools: - read_file - run_shell_command --- WGCNA-style weighted gene co-expression network analysis. Detects gene modules via soft thresholding, topological overlap, and hierarchical clust...