Walk through your analysis

Public accessions#

Start with a public RNA-seq study and follow the read-to-results route. No read-processing stage is skipped; the workflow first retrieves the reads.

1. Prepare the environment: Make the tools required by your route available.
2. Fetch public metadata: Resolve the selected public study into its reads and sample records.
3. Review sample metadata: Connect observations to biological groups and relevant covariates.
4. Match the reference and identifiers: Choose the correct organism and annotation context.
5. Choose read processing: Define how reads become gene-level measurements.
6. Define the comparison: Specify the biological question and selected-gene criteria.
7. Choose interpretation outputs: Select enrichment, custom sets and network analysis that fit the study.
8. Set resources: Give the workflow a feasible local budget.
9. Validate and inspect the plan: Catch inconsistent inputs before expensive work.
10. Run and monitor: Execute the validated route.
11. Read and export results: Interpret the outputs and preserve their provenance.

Local FASTQ#

Start with reads already available on disk. Public read download is skipped. Read QC, processing and quantification still apply.

1. Prepare the environment: Make the tools required by your route available.
2. Select local FASTQ: Connect reads already on disk to the project.
3. Review sample metadata: Connect observations to biological groups and relevant covariates.
4. Match the reference and identifiers: Choose the correct organism and annotation context.
5. Choose read processing: Define how reads become gene-level measurements.
6. Define the comparison: Specify the biological question and selected-gene criteria.
7. Choose interpretation outputs: Select enrichment, custom sets and network analysis that fit the study.
8. Set resources: Give the workflow a feasible local budget.
9. Validate and inspect the plan: Catch inconsistent inputs before expensive work.
10. Run and monitor: Execute the validated route.
11. Read and export results: Interpret the outputs and preserve their provenance.

Raw count matrix#

Fit a local count model from unnormalized gene counts. Download, read QC, trimming, alignment and counting are skipped. No read-level QC is reconstructed.

1. Prepare the environment: Make the tools required by your route available.
2. Import raw counts: Start after read processing and counting.
3. Review sample metadata: Connect observations to biological groups and relevant covariates.
4. Match the reference and identifiers: Choose the correct organism and annotation context.
5. Define the comparison: Specify the biological question and selected-gene criteria.
6. Choose interpretation outputs: Select enrichment, custom sets and network analysis that fit the study.
7. Set resources: Give the workflow a feasible local budget.
8. Validate and inspect the plan: Catch inconsistent inputs before expensive work.
9. Run and monitor: Execute the validated route.
10. Read and export results: Interpret the outputs and preserve their provenance.

Microarray#

Analyse normalized intensities through the limma route. RNA-seq read processing and counting do not apply. Array processing and mapping checks replace those stages.

1. Prepare the environment: Make the tools required by your route available.
2. Import microarray intensities: Choose processed intensities or supported raw-array processing.
3. Review sample metadata: Connect observations to biological groups and relevant covariates.
4. Match the reference and identifiers: Choose the correct organism and annotation context.
5. Define the comparison: Specify the biological question and selected-gene criteria.
6. Choose interpretation outputs: Select enrichment, custom sets and network analysis that fit the study.
7. Set resources: Give the workflow a feasible local budget.
8. Validate and inspect the plan: Catch inconsistent inputs before expensive work.
9. Run and monitor: Execute the validated route.
10. Read and export results: Interpret the outputs and preserve their provenance.

Finished DE results#

Interpret an external model with its source direction preserved. Reads, alignment, counting and local model fitting are skipped. Without a sample-expression matrix there is no PCA, count heatmap or normalized-expression export.

1. Prepare the environment: Make the tools required by your route available.
2. Import finished statistics: Reuse a completed upstream model without fitting a local one.
3. Match the reference and identifiers: Choose the correct organism and annotation context.
4. Choose downstream thresholds: Select gene lists from supplied values without refitting.
5. Choose interpretation outputs: Select enrichment, custom sets and network analysis that fit the study.
6. Set resources: Give the workflow a feasible local budget.
7. Validate and inspect the plan: Catch inconsistent inputs before expensive work.
8. Run and monitor: Execute the validated route.
9. Read and export results: Interpret the outputs and preserve their provenance.