4. Quality Control

Aim:

• Assess the quality of raw datasets

• Define quality trimming parameters prior running the complete RNAseq gene profiling pipeline

Run RNA-seq QC check

The pipeline requires preparing at least 2 files:

• Metadata file (samplesheet.csv) that specifies the name of the samples, location of FASTQ files ('Read 1' and ‘Read 2’), and strandedness (forward, reverse, or auto. Note: auto is used when the strandedness of the data is unknown)

• PBS Pro script (launch_nf-core_RNAseq_QC.pbs) with instructions to run the pipeline

Create the metadata file (samplesheet.csv):

Change to the data folder directory:

Copy the bash script to the working folder

• Note: you could replace ‘$HOME/workshop/data’ with “.” A dot indicates ‘current directory’ and will copy the file to the directory where you are currently located

View the content of the script:

Example for Single-End data (when only ‘Read 1’ is available):

Open

NOTE: modify ‘read1_extension’ as appropriate for your data. For example: _1.fastq.gz or _R1_001.fastq.gz or _R1.fq.gz , etc

Let’s generate the metadata file by running the following command:

Check the newly created samplesheet.csv file:

Copy the PBS Pro script for QC (launch_nf-core_RNAseq_QC.pbs)

Copy and paste the code below to the terminal:

• Line 1: Copy the samplesheet.csv file to the working directory

• Line 2: move to the working directory

• Line 3: copy the launch_nf-core_RNAseq_QC.pbs submission script to the working directory

View the content of the launch_nf-core_RNAseq_QC.pbs script:

Open

• We recommend running the nextflow nf-core/rnaseq pipeline once and then assessing the fastqc results folder to assess if sequence biases are present in the 5'-end and 3'-end ends of the sequences.

• Version 3.12.0 allows running the pipeline to do quality assessment only, without any alignment, read counting, or trimming.

Submitting the job

Once you have created the folder for the run, the samplesheet.csv file, and launch.pbs, you are ready to submit the job to the HPC scheduler:

Monitoring the Run

to check on the jobs, you are running. Nextflow will launch additional jobs during the run.

You can also check the .nextflow.log file for details on what is going on.

Once the pipeline has finished running - Assess the QC report:

NOTE: To proceed, you need to be on QUT’s WiFi network or signed via VPN.

To browse the working folder in the HPC type in the file finder:

Windows PC

Mac

Evaluate the nucleotide distributions in the 5'-end and 3'-end of the sequenced reads (Read1 and Read2). Look into the “MultiQC” folder and open the provided HTML report.

Items to check:

• The overall quality of the experiment and reads. Look at the “Sequence Quality Histogram” plot. For example, if Phred assigns a quality score of 30 to a base, the chances that this base is called incorrectly are 1 in 1000. Phred quality scores are logarithmically linked to error probabilities.

Inspect precomputed QC reports

FastQC:

• location: /work/training/2024/rnaseq/runs/run2_QC/results/fastqc

• Quality control metrics to assess - an individual html report is produced for each sample:

  • Per Base Sequence Quality: Measures the quality score across all bases at each position in the reads, providing a visual summary of how the quality varies across the length of the reads.

  • Per Sequence Quality Scores: Shows the distribution of mean quality scores for all reads, highlighting if the dataset contains many low-quality reads.

  • Per Base Sequence Content: Evaluates the proportion of each nucleotide (A, T, C, G) at every base position, revealing any biases in nucleotide composition

  • Per Base GC Content: Measures the percentage of GC content per base, checking for consistency with the expected GC content for the species or sample type.

  • Per Sequence GC Content: Examines the overall distribution of GC content across all reads, helping to detect unusual GC bias or contamination

  • Per Base N Content: Reports the proportion of uncalled bases (“N”s) at each position in the reads, indicating sequencing or base-calling issues

  • Sequence Length Distribution: Displays the distribution of read lengths, which can help assess whether the sequencing run generated reads of the expected length.

  • Sequence Duplication Levels: Identifies the level of duplicate sequences in the dataset, which can indicate issues such as overamplification in PCR or technical biases.

  • Overrepresented Sequences: Detects sequences that appear more frequently than expected, which might indicate contamination, adapters, or other technical artifacts.

  • Adapter Content: Checks for the presence of sequencing adapters that were not properly removed during preprocessing, which can affect downstream analysis.

  • K-mer Content: Identifies enriched k-mers (short nucleotide sequences) that occur more frequently than expected, possibly indicating contamination or other biases.

• MultiQC:

  • Combined FastQC results for all samples:

    • Per Base Sequence Quality: Overview of quality scores across bases.

    • Per Sequence Quality Scores: Mean quality scores of reads.

    • Per Base Sequence Content: Distribution of nucleotide composition across reads.

    • Per Sequence GC Content: GC content distribution.

    • Sequence Length Distribution: The length of reads in the data.

    • Adapter Content: Detection of sequencing adapters.

Hit: define how many nucleotides should be trimmed from the 5'-end and/or 3-end regions of the FASTQ reads.