Metagenomics Sequence Processing

Metagenomic raw sequences consist of strings composed of four letters: A, C, T, and G, representing the nucleotide bases Adenine, Cytosine, Thymine, and Guanine, respectively. These bases form the building blocks of DNA. Machines that sequence DNA extract these bases from a DNA sample and store them in files, commonly in formats such as FASTQ or FASTA.

In this tutorial, we will focus on the widely adopted FASTQ format. A FASTQ file typically contains essential information about the sample, the nucleotide sequences, and a quality score for each identified base. The quality score indicates the probability of correctly identifying a specific base, providing a valuable measure of sequencing accuracy.

We will use FASTQ files to prepare data for further analysis. The figure below illustrates the processing pipeline and the tools involved in this workflow.

flowchart LR
    A(FASTQ) --> B(Quality Control)
    B --> I(Host DNA Removal)
    I --> C(Sequence Assembly)
    C --> D(Taxonomic Profiling)
    C --> E(Functional Annotation)
    D --> F(Taxonomic Profile)
    E --> G(Functional Pathways)

• Quality Control: This is the first step where we will assess the quality of raw sequences to identify low-quality reads and adapter contamination. Tools such as FASTQC, MULTIQC, and Fastp will help generate quality metrics like per-based quality scores and read length distribution. Trimming tools like Trimmomatic or Trim-glore then help clean the data.

• Human DNA Removal: This step removes host DNA from metagenomic raw sequences to ensure that the downstream analysis focuses fully to microbial sequences. Tools such as bowtie2 and samtools will be used for this task. The reference human genome (e.g., GRCh38) will be downloaded and indexed before alignment.

• Sequence Assembly: In this step, we will assemble short reads into longer sequences. Tools like SPAdes (ideal for small database) and MegaHit (optimized for large datasets) will be employed for assembly.

• Taxonomic Profiling: This step assigns operational taxonomic units (OTUs) to assembeled and quality controled raw sequences to determine the taxonomic composition of the microbiome. To execute this task, we will explore the usage of tools such as MetaPhlan2, BowTie2, and mOTUs.

• Functional Annotation: While taxonomic profile identifies “who is there”, functional annotation determines “what they are doing”. It assigns functional roles and metabolic pathways present in provide metagenomic sequences. Tools like HUMAnN will be used for the task.