RNA-seq Analysis Pipeline

This repository contains RNA-seq analysis workflows for MAPK/ERK pathway inhibitor studies in lung cancer cell lines. Features publication-quality GSEA visualizations and transcription factor family enrichment analysis.

Projects

Project	Cell Line	Treatments	Description
8442	PC9 (EGFR-mutant NSCLC)	DMSO, Gefitinib (EGFR-i), SCH772984 (ERK-i)	EGFR vs ERK inhibition
8446	A549 (KRAS-mutant NSCLC)	DMSO, LY300 (RAF-i), SCH772984 (ERK-i)	RAF vs ERK inhibition

Key Features

Publication-quality GSEA dot plots with viridis color scale
Four pathway databases: Hallmark, KEGG, Reactome, GO:BP
Comparison plots with NES-based color encoding (orangered=up, royalblue=down)
Combined 4-panel figures for cross-database comparison
Transcription Factor family enrichment using CISBP database (~60 DBD families)

Directory Structure

RNA-seq_pipeline/
├── README.md
├── 8442_PC9_analysis/           # PC9 cell line analysis
│   ├── 20251218_Set_up_files_for_nf-core.Rmd
│   ├── 20251220_RNA-seq_analysis.Rmd
│   ├── 20260120_RNA-seq_analysis.Rmd          # Updated GSEA visualizations
│   ├── 20260120_RNA-seq_analysis_TF_enrichment.Rmd  # TF family analysis
│   ├── setup_rnaseq_pipeline.sh
│   ├── run_nfcore_rnaseq_ULTRA_SAFE.sh
│   └── launch_pipeline_ultra_safe.sh
│
└── 8446_A549_analysis/          # A549 cell line analysis
    ├── 20260106_Set_up_files_for_nf-core.Rmd
    ├── 20260106_RNA-seq_analysis.Rmd
    ├── setup_rnaseq_pipeline.sh
    ├── run_nfcore_rnaseq_ULTRA_SAFE.sh
    └── launch_pipeline_ultra_safe.sh

Workflow Overview

The analysis workflow consists of two main phases:

Phase 1: Data Processing (nf-core/rnaseq)

Data Ingestion (*_Set_up_files_for_nf-core.Rmd)
- Parse FASTQ filenames to extract metadata
- Create sample sheet for nf-core pipeline
- Inspect file sizes and estimate read counts
- Export metadata and sample sheet CSV
Pipeline Setup (setup_rnaseq_pipeline.sh)
- Validate prerequisites (conda, R, tidyverse)
- Create conda environment with Nextflow
- Generate nf-core sample sheet format
- Create pipeline execution scripts
Pipeline Execution (run_nfcore_rnaseq_ULTRA_SAFE.sh or launch_pipeline_ultra_safe.sh)
- Run nf-core/rnaseq with STAR + Salmon
- Ultra-safe mode: 1 STAR job at a time, 28GB RAM limit
- Automatic resume on failure
- ~10-12 hours runtime for 9 samples

Phase 2: Differential Expression Analysis

DESeq2 Analysis (*_RNA-seq_analysis.Rmd)
- Load counts from Salmon output
- Quality control (PCA, sample distances, correlations)
- Differential expression analysis
- Volcano plots, MA plots, heatmaps
- Gene Set Enrichment Analysis (fgsea) with 4 databases:
  - MSigDB Hallmark (50 gene sets)
  - KEGG pathways
  - Reactome pathways
  - GO Biological Process
- Publication-quality GSEA dot plots
- Cross-database comparison figures
TF Family Enrichment (*_TF_enrichment.Rmd) - NEW
- Transcription factor family enrichment using CISBP database
- ~60 DNA-binding domain (DBD) families analyzed
- Individual plots per treatment + comparison plots
- Shows TFs with coordinated expression changes per family

Quick Start Guide

Step 1: Run Data Ingestion (Local R/RStudio)

Open the *_Set_up_files_for_nf-core.Rmd file in RStudio and knit to HTML. This will:

Parse FASTQ filenames
Create sample_sheet.csv
Generate metadata files

Step 2: Copy Sample Sheet to Server

# For 8442 (PC9)
scp sample_sheet.csv user@server:/DATA/usr/d.goodall/Projects/PARM/E2845_PC9_Gef30nM_Sch810nM/RNA_seq/8442_PC9_analysis/

# For 8446 (A549)
scp sample_sheet.csv user@server:/DATA/usr/d.goodall/Projects/PARM/E2848_A549_Ly300_Sch77/RNA-seq/8446_A549_analysis/

Step 3: Run Pipeline Setup (Server)

# Navigate to analysis directory
cd /DATA/usr/d.goodall/Projects/PARM/E2848_A549_Ly300_Sch77/RNA-seq/8446_A549_analysis

# Run setup script
bash setup_rnaseq_pipeline.sh

Step 3b: Make Scripts Executable (if needed)

If you get "Permission denied" errors when running scripts, make them executable:

# Make all shell scripts executable
chmod +x *.sh

# Or individually:
chmod +x launch_pipeline_ultra_safe.sh
chmod +x run_nfcore_rnaseq_ULTRA_SAFE.sh
chmod +x setup_rnaseq_pipeline.sh

Step 4: Launch Pipeline (Server)

Option A: Background execution (recommended)

bash launch_pipeline_ultra_safe.sh

Option B: Interactive execution

bash run_nfcore_rnaseq_ULTRA_SAFE.sh

Option C: SLURM submission

sbatch --cpus-per-task=16 --mem=64G --time=24:00:00 \
       --job-name=rnaseq_8446 run_nfcore_rnaseq_ULTRA_SAFE.sh

Step 5: Monitor Progress

# View log
tail -f pipeline_ultra_safe.log

# Check STAR jobs (should be 0 or 1)
ps aux | grep "STAR --genomeDir" | grep -v grep | wc -l

# Check memory usage
free -h

if need to kill the processes

1. Find your processes:

ps aux | grep $(whoami) | grep -iE "nextflow|java|star|salmon|trim_galore|fastqc|samtools|picard" | grep -v grep
2. Kill them all at once:


# Kill nextflow and its spawned Java process first
pkill -u $(whoami) -f nextflow
pkill -u $(whoami) -f "java.*nextflow"

# Kill any remaining pipeline tools
pkill -u $(whoami) -f STAR
pkill -u $(whoami) -f salmon
pkill -u $(whoami) -f trim_galore
pkill -u $(whoami) -f fastqc
pkill -u $(whoami) -f samtools
pkill -u $(whoami) -f picard

Before re-running on HPC, clean the failed task caches:

cd /DATA/shared/projects/EGFR_TFs/00_MAPK/01_PARM/03_E2896_H1975_Osi_Sch77/RNA-seq
conda activate rnaseq
nextflow clean -f

The below checks if we are good to perform a re-run

#!/bin/bash

Pre-run diagnostic check for nf-core/rnaseq re-run

ANALYSIS_DIR="/DATA/shared/projects/EGFR_TFs/00_MAPK/01_PARM/03_E2896_H1975_Osi_Sch77/RNA-seq"

echo "═══════════════════════════════════════════════════" echo " PRE-RUN DIAGNOSTIC CHECK" echo "═══════════════════════════════════════════════════" echo ""

1. Check for running processes

echo "1. RUNNING PROCESSES" NF_PROCS=$(ps aux | grep -E "nextflow|java.*nextflow" | grep -v grep | wc -l) STAR_PROCS=$(ps aux | grep "STAR" | grep -v grep | wc -l) echo " Nextflow processes: ${NF_PROCS}" echo " STAR processes: ${STAR_PROCS}" [ "$NF_PROCS" -eq 0 ] && [ "$STAR_PROCS" -eq 0 ] && echo " ✅ All clear" || echo " ❌ Kill these first!" echo ""

2. Check nextflow clean status

echo "2. NEXTFLOW WORK DIR" if [ -d "${ANALYSIS_DIR}/work" ]; then WORK_SIZE=$(du -sh "${ANALYSIS_DIR}/work" 2>/dev/null | cut -f1) WORK_DIRS=$(find "${ANALYSIS_DIR}/work" -maxdepth 2 -mindepth 2 -type d 2>/dev/null | wc -l) echo " Work dir exists: yes (${WORK_SIZE}, ${WORK_DIRS} task dirs)" echo " ⚠️ Run 'nextflow clean -f' if not done yet" else echo " Work dir: cleaned or not present" echo " ✅ Clean" fi echo ""

3. Check lock files

echo "3. LOCK FILES" if [ -f "${ANALYSIS_DIR}/.nextflow/lock" ]; then echo " ❌ Lock file exists: ${ANALYSIS_DIR}/.nextflow/lock" echo " Run: rm ${ANALYSIS_DIR}/.nextflow/lock" else echo " ✅ No lock file" fi echo ""

4. Check report files that caused overwrite error

echo "4. EXISTING REPORT FILES" REPORT="${ANALYSIS_DIR}/results_8522/pipeline_info/execution_report_ultra_safe.html" TIMELINE="${ANALYSIS_DIR}/results_8522/pipeline_info/timeline_ultra_safe.html" [ -f "$REPORT" ] && echo " Report exists: yes (will be overwritten with new config)" || echo " Report: not present" [ -f "$TIMELINE" ] && echo " Timeline exists: yes (will be overwritten with new config)" || echo " Timeline: not present" echo " ✅ OK - new config has overwrite = true" echo ""

5. Check conda env

echo "5. CONDA ENVIRONMENT" echo " Active env: ${CONDA_DEFAULT_ENV:-none}" if command -v nextflow &> /dev/null; then NF_VER=$(nextflow -version 2>&1 | grep -oP 'version \K[0-9.]+' || echo "unknown") echo " Nextflow version: ${NF_VER}" echo " ✅ Nextflow available" else echo " ❌ Nextflow not found - activate conda env first" fi echo ""

6. Check sample sheet

echo "6. SAMPLE SHEET" SS="${ANALYSIS_DIR}/nfcore_samplesheet.csv" if [ -f "$SS" ]; then SAMPLES=$(tail -n +2 "$SS" | wc -l) echo " Found: ${SS}" echo " Samples: ${SAMPLES}" echo " ✅ Ready" else echo " ❌ Not found: ${SS}" fi echo ""

7. Check cached results

echo "7. CACHED RESULTS (from previous run)" COUNT_FILE="${ANALYSIS_DIR}/results_8522/star_salmon/salmon.merged.gene_counts.tsv" [ -f "$COUNT_FILE" ] && echo " Gene counts file: exists ✅" || echo " Gene counts file: not yet" STAR_BAMS=$(find "${ANALYSIS_DIR}/results_8522/star_salmon" -name "*.bam" 2>/dev/null | wc -l) echo " STAR BAM files: ${STAR_BAMS}" echo ""

8. Memory check

echo "8. SYSTEM RESOURCES" AVAIL_GB=$(free -g | grep Mem | awk '{print $7}') echo " Available memory: ${AVAIL_GB} GB" [ "$AVAIL_GB" -ge 50 ] && echo " ✅ Sufficient (need ~50GB peak)" || echo " ⚠️ Low - need ~50GB peak" echo ""

9. Check the updated script is in place

echo "9. UPDATED RUN SCRIPT" SCRIPT="${ANALYSIS_DIR}/run_nfcore_rnaseq_ULTRA_SAFE.sh" if [ -f "$SCRIPT" ]; then if grep -q "report.overwrite" "$SCRIPT" 2>/dev/null; then echo " ✅ Script has report.overwrite fix" else echo " ❌ Script is OLD version - copy updated script from repo" fi if grep -q "max_memory = '40.GB'" "$SCRIPT" 2>/dev/null; then echo " ✅ Script has updated max_memory (40GB)" else echo " ❌ Script has old max_memory - copy updated script from repo" fi else echo " ❌ Script not found at ${SCRIPT}" fi echo ""

echo "═══════════════════════════════════════════════════" echo " DONE - Review above and fix any ❌ items" echo "═══════════════════════════════════════════════════"

Check which sample is processing

ps aux | grep "STAR --genomeDir" | grep -v grep

Check STAR jobs (MOST IMPORTANT): ps aux | grep "STAR --genomeDir" | grep -v grep | wc -l

See if still running

ps aux | grep nextflow | grep -v grep

Step 6: Run DESeq2 Analysis (Local R/RStudio)

After pipeline completes (~10-12 hours), open *_RNA-seq_analysis.Rmd in RStudio and knit to HTML.

Pipeline Configuration

Resource Limits (Ultra-Safe Mode)

Process	Max Forks	Memory	CPUs
STAR Alignment	1	28 GB	16
Picard MarkDuplicates	2	12 GB	-
Salmon Quant	2	6 GB	-
RSeQC	2	8 GB	-

Expected Runtime

Per STAR job: ~45-60 minutes
9 samples: ~9 hours
Post-processing: ~1 hour
Total: ~10-12 hours

Output Files

From nf-core/rnaseq

results_*/
├── star_salmon/
│   ├── salmon.merged.gene_counts.tsv    # Gene-level counts
│   ├── salmon.merged.gene_tpm.tsv       # TPM values
│   └── [sample]/                        # Per-sample Salmon output
├── multiqc/
│   └── multiqc_report.html              # QC summary
├── fastqc/                              # Raw read QC
├── trimgalore/                          # Trimmed reads
└── pipeline_info/                       # Execution reports

From DESeq2 Analysis

DESeq2_analysis/
├── PCA_plot.png
├── sample_distance_heatmap.png
├── correlation_heatmap.png
├── dispersion_plot.png
├── volcano_plot_*.png
├── MA_plots_combined.png
├── heatmap_top_DEGs.png
├── heatmap_MAPK_pathway.png
├── DESeq2_results_*.csv                 # Full results tables
│
├── # GSEA Results (CSV)
├── GSEA_Hallmark_*.csv
├── GSEA_KEGG_*.csv
├── GSEA_Reactome_*.csv
├── GSEA_GOBP_*.csv                      # NEW: GO Biological Process
│
├── # GSEA Dot Plots (PNG + PDF)
├── GSEA_Hallmark_dotplot_*.{png,pdf}    # Individual treatment plots
├── GSEA_KEGG_dotplot_*.{png,pdf}
├── GSEA_Reactome_dotplot_*.{png,pdf}
├── GSEA_GOBP_dotplot_*.{png,pdf}
│
├── # GSEA Comparison Plots
├── GSEA_Hallmark_comparison_dotplot.{png,pdf}
├── GSEA_KEGG_comparison_dotplot.{png,pdf}
├── GSEA_Reactome_comparison_dotplot.{png,pdf}
├── GSEA_GOBP_comparison_dotplot.{png,pdf}
├── GSEA_ALL_databases_comparison.{png,pdf}  # 4-panel combined figure
│
├── # TF Family Enrichment
├── TF_family_dotplot_Gefitinib.{png,pdf}
├── TF_family_dotplot_SCH772984.{png,pdf}
├── TF_family_comparison_dotplot.{png,pdf}
├── GSEA_TF_families_*.csv
│
├── # Individual TF Gene Analysis (NEW)
├── TF_individual_dotplot_Gefitinib.{png,pdf}
├── TF_individual_dotplot_SCH772984.{png,pdf}
├── TF_individual_comparison_dotplot.{png,pdf}  # 20" x 14" faceted
│
├── DEG_summary_table.csv
└── GSEA_summary_table.csv

Experimental Details

Project 8442: PC9 (EGFR-mutant)

Cell line: PC9 (EGFR exon 19 deletion)
Treatments:
- DMSO (vehicle control)
- Gefitinib 30 nM (EGFR inhibitor, 24h)
- SCH772984 810 nM (ERK1/2 inhibitor, 24h)
Replicates: 3 biological replicates (B1, B2, B3)
FASTQ path: /shared/gcf/d.goodall/8442/fastq_files
Output path: /DATA/usr/d.goodall/Projects/PARM/E2845_PC9_Gef30nM_Sch810nM/RNA_seq/8442_PC9_analysis

Project 8446: A549 (KRAS-mutant)

Cell line: A549 (KRAS G12S mutation)
Treatments:
- DMSO (vehicle control)
- LY300 (RAF inhibitor)
- SCH772984 (ERK1/2 inhibitor)
Replicates: 3 biological replicates (B1, B2, B3)
FASTQ path: /shared/gcf/d.goodall/8446/fastq_files
Output path: /DATA/usr/d.goodall/Projects/PARM/E2848_A549_Ly300_Sch77/RNA-seq/8446_A549_analysis

Dependencies

R Packages

# Core analysis
library(tidyverse)
library(DESeq2)
library(AnnotationDbi)
library(org.Hs.eg.db)

# Visualization
library(ggplot2)
library(ggrepel)
library(pheatmap)
library(RColorBrewer)
library(ComplexHeatmap)
library(EnhancedVolcano)
library(scales)           # For color scaling

# GSEA
library(fgsea)
library(msigdbr)
library(pathview)

# Utilities
library(biomaRt)
library(ggpubr)
library(apeglm)
library(patchwork)        # Multi-panel figures

# TF Enrichment (NEW)
library(TFutils)          # CISBP TF catalog with DBD families

Command-line Tools

Nextflow (via conda)
nf-core/rnaseq v3.17.0 (Note: requires process.shell fix for versions 3.16.1-3.18.0)
Singularity (recommended) or Docker for containers (more reliable than conda)

Troubleshooting

Pipeline Fails

# Check Nextflow log
tail -100 .nextflow.log

# Check for errors
grep -i error .nextflow.log

# Resume from last checkpoint
bash launch_pipeline_ultra_safe.sh  # -resume flag is included

CRLF Line Ending Errors (Windows → Linux)

If you see errors like:

/bin/bash^M: bad interpreter: No such file or directory
'\r': command not found

Cause: Scripts edited on Windows have CRLF line endings instead of Unix LF.

Fix:

# Convert all shell scripts to Unix line endings
sed -i 's/\r$//' *.sh

# Or use dos2unix if available
dos2unix *.sh

Prevention: The repository includes a .gitattributes file that enforces LF line endings for shell scripts.

process.shell Permission Denied (Exit Code 126)

If you see errors like:

.command.run: line 330: .command.run: Permission denied
Command exit status: 126
WARN: Directive `process.shell` cannot contain new-line characters

Cause: Known bug in nf-core/rnaseq versions 3.16.1, 3.17.0, and 3.18.0. The process.shell directive contains newline characters that break newer Nextflow versions.

Fix: Already applied in run_nfcore_rnaseq_ULTRA_SAFE.sh. The custom config includes:

process.shell = ['/bin/bash', '-euo', 'pipefail']

References:

Conda Package Not Found (fq, etc.)

If you see errors like:

PackagesNotFoundError: bioconda::fq==0.9.1
PackagesNotFoundError: bioconda::fq==0.12.0

Cause: The fq package is only needed when strandedness = "auto" in the sample sheet, which triggers the FQ_SUBSAMPLE step.

Fix (already applied): The setup script now uses strandedness = "reverse" instead of "auto":

mutate(strandedness = "reverse")  # Avoids fq package requirement

Quick fix if you already have a sample sheet:

sed -i 's/,auto$/,reverse/' nfcore_samplesheet.csv

Alternative solutions:

Use Singularity instead of Conda (recommended for better reproducibility):

# Edit run_nfcore_rnaseq_ULTRA_SAFE.sh
# Change: -profile conda
# To:     -profile singularity

Use Docker instead of Conda:

# Change: -profile conda
# To:     -profile docker

Out of Memory (OOM)

The ultra-safe configuration limits STAR to 28GB RAM with only 1 job at a time. If still failing:

Check for other processes using memory: free -h
Reduce --max_memory in config
Contact system admin

Disk Space Issues

If you see errors like:

No space left on device
Unable to stage foreign file: s3://ngi-igenomes/...

Check disk space:

df -h /DATA

Free up space:

# Delete work directories from previous runs (ONLY after pipeline completes!)
rm -rf work/

# Check large files in your directory
find /DATA/usr/yourusername/ -type f -size +1G -exec ls -lh {} \; 2>/dev/null | head -20

Note: The STAR index download from S3 requires ~30GB of free space. The work directory can grow to 100+ GB during execution.

Sample Sheet Issues

Ensure sample sheet format matches:

sample,fastq_1,fastq_2,strandedness
A549_dmso_B1,/path/to/R1.fastq.gz,/path/to/R2.fastq.gz,reverse

Important: Use reverse for strandedness (standard for Illumina stranded RNA-seq) to avoid the fq package issue.

GSEA Visualization Details

Individual Dot Plots

Publication-quality dot plots for each pathway database:

Aesthetic	Encoding
X-axis	NES (Normalized Enrichment Score)
Y-axis	Pathway name (ordered by NES)
Dot size	Gene count in pathway
Dot color	-log10(padj) using viridis scale

Vertical dashed line at NES = 0
Top 30 pathways shown per database
Significance threshold: padj < 0.05

Comparison Dot Plots

Side-by-side comparison of treatments:

Aesthetic	Encoding
X-axis	Treatment (Gefitinib, SCH772984)
Y-axis	Pathway name (clustered by NES pattern)
Dot color	NES direction: `orangered` (up) / `royalblue` (down)
Dot alpha	NES magnitude (0.3-1.0)
Dot size	Gene count
Markers	`*` padj ≤ 0.01, `.` padj ≤ 0.05

Combined 4-Panel Figure

GSEA_ALL_databases_comparison.{png,pdf} combines all four database comparisons:

┌─────────────────┬─────────────────┐
│  A. Hallmark    │  B. KEGG        │
├─────────────────┼─────────────────┤
│  C. Reactome    │  D. GO:BP       │
└─────────────────┴─────────────────┘

16" × 20" portrait format
Panel labels (A-D) for publication reference
Unified title and legend

TF Family Enrichment Analysis

Uses the CISBP database via TFutils to analyze transcription factor families:

Method

TF genes grouped by DNA-binding domain (DBD) family (~60 families)
fgsea enrichment analysis on ranked gene lists
Leading edge size = TFs actually contributing to enrichment signal

Key Parameters

Parameter	Value	Rationale
`minSize`	2	Capture small TF families
`padj_cutoff`	0.25	Exploratory threshold
`min_tfs_changing`	1	Capture individual TF changes
Size breaks	1, 2, 3, 4, 5+	Leading edge counts

TF Family Plots

Individual dot plots per treatment (NES on x-axis, viridis color for significance)
Comparison plot with - indicator for TF families not significant in one treatment
Same orangered/royalblue encoding as GSEA comparison plots

Individual TF Gene Plots (NEW)

Shows actual TF gene names (not grouped by family):

Feature	Value
Ranking	By absolute log2 fold change
Filter	padj < 0.25
Display	Top 100 TFs per treatment
Comparison	Side-by-side faceted layout (20" × 14")

Output Files

File	Description
`TF_family_dotplot_*.{png,pdf}`	Family-level enrichment per treatment
`TF_family_comparison_dotplot.{png,pdf}`	Side-by-side family comparison
`TF_individual_dotplot_*.{png,pdf}`	Individual TF genes per treatment
`TF_individual_comparison_dotplot.{png,pdf}`	Faceted individual TF comparison
`GSEA_TF_families_*.csv`	Full enrichment results

References

nf-core/rnaseq
DESeq2 vignette
fgsea
TFutils - TF family classification
MSigDB - Pathway gene sets

Author

D.J. Goodall

License

This project is for internal research use.

Name		Name	Last commit message	Last commit date
8442_PC9_analysis		8442_PC9_analysis
8446_A549_analysis		8446_A549_analysis
8522_H1975_analysis		8522_H1975_analysis
.gitattributes		.gitattributes
.gitignore		.gitignore
README.md		README.md
RNAseq_SOP.md		RNAseq_SOP.md

Latest commit

History

Folders and files

Repository files navigation

RNA-seq Analysis Pipeline

Projects

Key Features

Directory Structure

Workflow Overview

Phase 1: Data Processing (nf-core/rnaseq)

Phase 2: Differential Expression Analysis

Quick Start Guide

Step 1: Run Data Ingestion (Local R/RStudio)

Step 2: Copy Sample Sheet to Server

Step 3: Run Pipeline Setup (Server)

Step 3b: Make Scripts Executable (if needed)

Step 4: Launch Pipeline (Server)

Step 5: Monitor Progress

if need to kill the processes

Before re-running on HPC, clean the failed task caches:

The below checks if we are good to perform a re-run

Pre-run diagnostic check for nf-core/rnaseq re-run

1. Check for running processes

2. Check nextflow clean status

3. Check lock files

4. Check report files that caused overwrite error

5. Check conda env

6. Check sample sheet

7. Check cached results

8. Memory check

9. Check the updated script is in place

Check which sample is processing

See if still running

Step 6: Run DESeq2 Analysis (Local R/RStudio)

Pipeline Configuration

Resource Limits (Ultra-Safe Mode)

Expected Runtime

Output Files

From nf-core/rnaseq

From DESeq2 Analysis

Experimental Details

Project 8442: PC9 (EGFR-mutant)

Project 8446: A549 (KRAS-mutant)

Dependencies

R Packages

Command-line Tools

Troubleshooting

Pipeline Fails

CRLF Line Ending Errors (Windows → Linux)

process.shell Permission Denied (Exit Code 126)

Conda Package Not Found (fq, etc.)

Out of Memory (OOM)

Disk Space Issues

Sample Sheet Issues

GSEA Visualization Details

Individual Dot Plots

Comparison Dot Plots

Combined 4-Panel Figure

TF Family Enrichment Analysis

Method

Key Parameters

TF Family Plots

Individual TF Gene Plots (NEW)

Output Files

References

Author

License

About

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages