Windows systems, compressed files can be uncompressed with a
tool such as 7-Zip, while on macOS, one can simply right click on
the file and select “Uncompress”.
3.5.2 Demultiplexing
If multiple samples were run on a single flow cell for a given
instrument, each sample received a unique identifying adapter
(an “index” sequence) during library prep. Most modern sequencing instruments will automatically sort reads into FASTQ files per
index sequence (a process called “demultiplexing”), but some
instruments require this be done by the user. For example, Nanopore instruments did not introduce automatic demultiplexing after
a sequencing run until mid-2019.
3.5.3 Applying Quality
Control
After a FASTQ file per sample sequenced has been obtained (or a
pair for Illumina paired-end sequencing), best practices recommend the application of quality control measures to ensure a quality
assembled genome. While not explicitly necessary, assembly without quality control corrections is likely to lead to subpar results at
best; in the worst case, the resulting genome could be severely
misleading. Before genome assembly, FASTQ files must be parsed
and corrected for three potential sources of error: potentially erroneous basecalls (basecalls with low quality scores), artificial index
and adapter bases being included in reads, and reads not from the
sample of interest (either due to cross-contamination between
multiplexed samples or from environmental contamination).
3.5.4 Detecting
and Removing Low-Quality
Bases and Adapter
Sequences
The first step to ensuring a clean FASTQ file is to quantify the
degree to which reads in your FASTQ contain either low quality
bases or adapter sequences. Tools that detect these sequences vary
depending on the sequencing technology employed. For Illumina,
the tool of choice is FastQC [30]; while there are multiple options
for Nanopore, a recommended tool is NanoQC [31]. Although a
tool like FastQC may determine a FASTQ file needs no additional
quality trimming, this is relatively uncommon; bioinformatics best
practices dictate running tools to remove these sequences regardless, since doing so is fast and computationally inexpensive. Like
tools for quantifying the presence of problematic sequences, tools
for their removal also vary by sequencing instrument. Illumina
benefits from options for removal of these sequences built into
the instrument itself before creation of the initial FASTQ files.
Additional tools for filtering and trimming Illumina FASTQ files
include fastp [32], Trimmomatic [33], and cutadapt [34]. Tools for
filtering and trimming Nanopore FASTQ files include Nanofilt
[31] and Filtlong [35]. Briefly, these tools usually implement
removal of adapter sequences (“trimming”) by comparing read
sequences against a database of known sequencing adapters/
indexes and implement removal of low-quality base pairs from the
Mycobacterial ID and Characterization Using WGS
413
tool such as 7-Zip, while on macOS, one can simply right click on
the file and select “Uncompress”.
3.5.2 Demultiplexing
If multiple samples were run on a single flow cell for a given
instrument, each sample received a unique identifying adapter
(an “index” sequence) during library prep. Most modern sequencing instruments will automatically sort reads into FASTQ files per
index sequence (a process called “demultiplexing”), but some
instruments require this be done by the user. For example, Nanopore instruments did not introduce automatic demultiplexing after
a sequencing run until mid-2019.
3.5.3 Applying Quality
Control
After a FASTQ file per sample sequenced has been obtained (or a
pair for Illumina paired-end sequencing), best practices recommend the application of quality control measures to ensure a quality
assembled genome. While not explicitly necessary, assembly without quality control corrections is likely to lead to subpar results at
best; in the worst case, the resulting genome could be severely
misleading. Before genome assembly, FASTQ files must be parsed
and corrected for three potential sources of error: potentially erroneous basecalls (basecalls with low quality scores), artificial index
and adapter bases being included in reads, and reads not from the
sample of interest (either due to cross-contamination between
multiplexed samples or from environmental contamination).
3.5.4 Detecting
and Removing Low-Quality
Bases and Adapter
Sequences
The first step to ensuring a clean FASTQ file is to quantify the
degree to which reads in your FASTQ contain either low quality
bases or adapter sequences. Tools that detect these sequences vary
depending on the sequencing technology employed. For Illumina,
the tool of choice is FastQC [30]; while there are multiple options
for Nanopore, a recommended tool is NanoQC [31]. Although a
tool like FastQC may determine a FASTQ file needs no additional
quality trimming, this is relatively uncommon; bioinformatics best
practices dictate running tools to remove these sequences regardless, since doing so is fast and computationally inexpensive. Like
tools for quantifying the presence of problematic sequences, tools
for their removal also vary by sequencing instrument. Illumina
benefits from options for removal of these sequences built into
the instrument itself before creation of the initial FASTQ files.
Additional tools for filtering and trimming Illumina FASTQ files
include fastp [32], Trimmomatic [33], and cutadapt [34]. Tools for
filtering and trimming Nanopore FASTQ files include Nanofilt
[31] and Filtlong [35]. Briefly, these tools usually implement
removal of adapter sequences (“trimming”) by comparing read
sequences against a database of known sequencing adapters/
indexes and implement removal of low-quality base pairs from the
Mycobacterial ID and Characterization Using WGS
413