Quality Control and Preprocessing of Sequencing Reads

Equipment

1. Computer (OS: Linux-branches, such as Centos and Ubuntu. We recommend at least 16GB RAM and multiple cores)

Software

1. FastQC (Andrews, 2014)

   FastQC is a program designed to spot potential problems in high throughput sequencing datasets. If java is not installed, you can add it by doing the following:

   Ubuntu: sudo apt install default-jre

   The software is available for download at https://codeload.github.com/s-andrews/FastQC/zip/refs/heads/master (last accessed date: 31/10/2021) or https://github.com/s-andrews/FastQC.git downloaded by git (terminal command: git clone https://github.com/s-andrews/FastQC.git) (last accessed date: 31/10/2021). To install FastQC, unzip the zip file. A wrapper script called ‘fastqc’ is included in the top level directory of FastQC installation. You may need to make this file executable: chmod 755 fastqc. If you have conda installed on your computer, the easy recommended way is:

   conda activate

   conda install fastqc
   
   

2. iTools (He et al., 2013)

   In this study, we use iTools to provide useful statistics of sequence data, which includes the Fqtools module to deal with fastq files. The Fqtools module provides multiple functions, as follows: a) summarizes the quality and amount of data, as well as the GC content; b) filters or trims the reads according to sequencing quality; c) removes reads contaminated with adapter sequences; and d) splits reads according to the index sequence. The software is available for download at https://github.com/BGI-shenzhen/Reseqtools/blob/master/iTools_Code20180520.tar.gz (last accessed date: 31/10/2021) and installation instructions are in the Install.Readme file.

3. Cutadapt (Martin, 2011)

   Cutadapt searches for the adapter sequences in all reads and removes them. The software is available for download at https://codeload.github.com/jamescasbon/cutadapt/zip/refs/heads/master (last accessed date: 31/10/2021), and the git clone command is: https://github.com/marcelm/cutadapt.git (last accessed date: 31/10/2021). Installation is done by: python setup.py install –user

   The easiest way to install cutadapt is to use pip on the command line:

   pip install –user –upgrade 
   
   

4. Fastp (Chen et al., 2018)

   A tool designed to provide fast all-in-one preprocessing for fastq files. The software is available for download at https://codeload.github.com/OpenGene/fastp/zip/refs/heads/master (last accessed date: 31/10/2021), and the git clone command is: git clone https://github.com/OpenGene/fastp.git (last accessed date: 31/10/2021). Installation instructions are in the README.md file. Another way to install fastp with conda:

   conda install -c bioconda fastp
   
   

5. FASTX (Gordon and Hannon, 2010)

   The FASTX-Toolkit is a collection of command line tools for short reads Fasta/FastQ files preprocessing. The software is available for download at: https://codeload.github.com/agordon/fastx_toolkit/zip/refs/heads/master (last accessed date: 31/10/2021), and the git clone command is: git clone https://github.com/agordon/ fastx_toolkit. git (last accessed date: 31/10/2021). Installation instructions are in the README file.

   
   

Input data

1. FastQ format
   This is the most widely used format in sequence analysis. The format contains more information than the fasta format, through integrating quality scores. Each sequence requires at least four lines:
   1. The first line is the sequence header which starts with an ‘@’.
   2. The second line is the sequence.
   3. The third line starts with ‘+’.
   4. The fourth line contains the quality scores.
      
      
      
2. We downloaded two transcriptome fastq data by wget. The SRR accession number or Project number could be described in published studies with SRA data uploaded into NCBI. The user can find the SRR accession number in the SRA database, based on Project number in NCBI, and download the fastq format data from EBI website (https://www.ebi.ac.uk) as alternative access. The user should search by the SRR number, and download links can be found on the result page. The users can record the download links of fastq files to download data in the terminal. In this study, the run accession numbers we used are SRR2061397 and SRR2061398. The sequencing platform is Illumina Hiseq 2000 paired end sequencing. The organism is Arabidopsis thaliana, and the method to download links is as follows (last accessed date: 15/7/2021):
   wget -c http://ftp.sra.ebi.ac.uk/vol1/fastq/SRR206/007/SRR2061397/SRR2061397_1.fastq.gz
   wget -c http://ftp.sra.ebi.ac.uk/vol1/fastq/SRR206/007/SRR2061397/SRR2061397_2.fastq.gz
   wget -c http://ftp.sra.ebi.ac.uk/vol1/fastq/SRR206/008/SRR2061398/SRR2061398_1.fastq.gz
   wget -c http://ftp.sra.ebi.ac.uk/vol1/fastq/SRR206/008/SRR2061398/SRR2061398_2.fastq.gz
   
   

Case study

FastQC provides a modular set of analyses, which can be used to give a quick impression of whether the data has any problems that you should be aware of before doing any further analysis. The main functions of FastQC are:

1. Import data from BAM, SAM, or FastQ files.
2. Providing a quick overview to tell you in which areas there may be problems.
3. Summary graphs and tables to quickly assess your data.
4. Export of result to an HTML-based permanent report.

FastQC can be run either as an interactive graphical application. Alternatively, you can run the program in a non-interactive way.

Result Interpretation

Here, we use two tools (FastQC and iTools) to perform quality control and three other tools (cutadapt, fastp, and FASTX) for preprocessing of sequencing reads. Command-line examples are provided to show the basic usages of these tools.

For the FastQC, an HTML report can be generated by running FastQC in non-interactive mode. It includes different modules, such as basic statistics, per base sequence quality, per sequence quality scores, per base sequence content, per sequence GC content, per base N content, sequence duplication levels, overrepresented sequences, and adapter content. A warning or failure icon on the modules indicates some kind of systematic problem. Users can interpret FastQC reports via the link (https://www.bioinformatics.babraham.ac.uk/projects/fastqc/Help/).

For iTools, a txt file is generated by running iTools in the command line. This file contains multiple information, such as GC content, base quality distribution, and read quality distribution. The percentage of reads with a quality score Q30 (99.9% base accuracy) of SRR2061397_1 and SRR2061397_1 is 93.83% and 88.24%, separately (Table 1). The quality score is important to evaluate the reliability of reads, and a high score represents high sequence quality.

Table 1. Read quality distribution generated by iTools.

For cutadapt, new fastq files are generated by removing adapters. Standard out shows the summary of reads processed: 0.2% of reads contain adapters, and 6.4% of reads are trimmed with low quality (Table 2). The reads satisfying ≥Q30 were considered in our analysis. Reads with low quality (<Q30) and adapters are trimmed.

Table 2. Summary of cutadapt output.

Fastp supports automatic adapter trimming and works faster than other preprocessing tools, such as Trimmomatic or Cutadapt. Fastp improves the read quality to some extent (Table 3). The total reads and total bases are provided before and after filtering. Q20 and Q30 quality scores for reads and bases are also shown. There are 18,655,502 reads passing the filter. Other reads with low quality and too many N are discarded.

Table 3. Summary of fastp output.

FASTX provides a fastx_clipper tool to remove adapters and low-quality reads. Here, iTools shows the quality distribution of fastq files filtered by fastx_clipper (Table 4). Compared with data before filter, the quality after filter improves it to some degree (Table 1 and Table 4).

Table 4. The read quality summary of fastx_clipper.

Discussion

Here, we displayed five tools for quality control and preprocessing. The examples in this study exemplified how to use these tools. Users can explore more complex usage after learning the basic commands. The detail information for input, output, and sample data can be accessed via the website GitHub (https://github.com/Bio-protocol/Bioinformatics-Recipes-for-Plant-Genomics/tree/master).

Notes

Data used in this study are available in published databases (NCBI/EBI), and users can download them freely via different ways. Results in this study were generated by the tools described above; no additional tool was used. As the common functions of tools, the users could choose part tools, such as FastQC for quality control and fastp for preprocessing. After preprocessing, users could check their data quality by running FastQC again. We provide two ways (web page download or git clone in terminal) to download the tools in GitHub. Before using conda, anaconda should be installed (https://repo.anaconda.com/archive/Anaconda3-2021.05-Linux-x86_64.sh). Run the file (Anaconda3-2021.05-Linux-x86_64.sh) to install anaconda in the terminal by command: sh Anaconda3-2021.05-Linux-x86_64.sh, it will be installed in a default place, and the user can activate the conda environment by command: conda activate. Then, users can ask conda to install the tools needed.

Acknowledgments

This work was funded by the National Science Foundation of China (No.31770333, No.31370329, and No.11631012), the Program for New Century Excellent Talents in University (NCET-12-0896), and the Fundamental Research Funds for the Central Universities (No. GK201403004). This paper describes protocols from https://www.bioinformatics.babraham.ac.uk/projects/fastqc/ (last accessed date: 31/10/2021) and other references (Gordon and Hannon, 2010; Martin, 2011; He et al., 2013; Chen et al., 2018).

Competing interests

The authors declare no competing interests.

References

1. Andrews, S. (2014). FastQC A Quality Control tool for High Throughput Sequence Data.
2. Bolger, A. M., Lohse, M. and Usadel, B. (2014). Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30(15): 2114-2120.
3. Chen, S., Zhou, Y., Chen, Y. and Gu, J. (2018). fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics 34(17): i884-i890.
4. Gordon, A. and Hannon, G. J. (2010). Fastx-toolkit. FASTQ/A short-reads pre-processing tools.
5. He, W., Zhao, S., Liu, X., Dong, S., Lv, J., Liu, D., Wang, J. and Meng, Z. (2013). ReSeq Tools: An integrated toolkit for large-scale next-generation sequencing based resequencing analysis. Genet Mol Res 12(4): 6275-6283.
6. Martin, M. (2011). Cutadapt removes adapter sequences from high-throughput sequencing reads. Embnet Journal 17(1).

Supplementary information

Data and code availability: All data and code have been deposited to GitHub: https://github.com/Bio-protocol/Bioinformatics-Recipes-for-Plant-Genomics.