-
Notifications
You must be signed in to change notification settings - Fork 4
Check list
epermal edited this page Jul 29, 2020
·
7 revisions
Once the pipeline has ended successfully, please check the multiqc_report.html file to see if there is any problem.
Here is non exhaustive list of sanity checks
- General statistics : percentage of aligned reads according to sample name
- FeatureCounts : Assigned vs Unassigned
- Alignment scores : According to mapper provided data Mapped vs Unmapped
- Bowtie 1 : rRNA mapped data should be low less than 10-20 % ???
- Cutadapt : Trimmed sequence length should be less than 30pb
- FastQC : Reads quality
- Sequence Quality Histograms : Curve has to be flat
- Per Sequence GC Content : Gives an insight of the GC composition for a given experiment. Usually around 50% of GC
- Per Base N Content: Gives an insight of the undetermined bases for a given experiment. Hopefully around 0% of N
Check RNAdiff report
- Table 1 : Check file IDs and labels
- Table 3 to X : Summary of the count tables
- Figure 1 : Check if mapped reads are similar or not
- Figure 2 : Check percentage of null counts (the dotted line) that will be set to NA (log fold change and p-value) in the result table
- Figure 3 : Check the denity of raw counts. They must follow the same pattern
- Figure 4 : Check the percentage of reads associated with the gene with the higher count. It should/could be the same in all replicate for one condition. Gives an insight if the highest count gene for each condition.
- Figure 5 : The sample with similar conditions should/could cluster together
- Figure 6 : First 2 component of the PCA : PC1 = X percentage of the variance. Usually what is expected is PC1 80% PC2 20% for a clear differential expression
- Figure 7 : Check if Boxplot comparison that illustrate normalisation is OK, They should be aligned after normalisation according to the median.
- Figure 8 :
- Figure 9 : Highest histogram bar should be around 0
- Figure 10 and 11 : MAplot and volcano plot showing diffrentialy expressed genes. The 0 axis cut the graph in two parts : Up regulated and down regulated
Results
- Create a folder to report results
- Copy in this folder : the pdf report from Deseq2 the tables end figures folder fron Deseq2 the multiqc html report Any other asked files
This pipeline is part of the Sequana project. If you use sequana_rnaseq, please consider citing us. Visit the How to cite ? section. You may also visit the pipeline page and star us.