Virus_Identification_Process

Softwares and Databases

Softwares

• Bowtie2: v2.5.4
• Megahit: v1.2.9
• Diamond: v2.1.8
• SeqKit: v2.9.0
• Palmscan: v2.0
• EMBOSS: v6.6.0.0

Databases

• SILVA_138.2_NR99
• VirusHost Protein Database

Steps

Part 0. Software installation & Database deployment

Installation of other Softwares

• Using conda or mamba (recommended):

  conda create -c bioconda -c conda-forge -n virus emboss fastp diamond seqkit megahit unzip gxx_linux-64 bowtie2 -y
  conda activate virus

Installation of Palmscan

• Using git:

  git clone https://github.com/MennoMens/palmscan.git
  cd palmscan/src
  make

Build the database of VirusHost protein

diamond makedb --in ./virushostdb.formatted.cds.faa -d ./db_diamond/virushostdb_protein

Build the database of rRNA

cat ./SILVA_138.2_LSURef_NR99_tax_silva.fasta ./SILVA_138.2_SSURef_NR99_tax_silva.fasta > SILVA_138.2_Ref_NR99.fasta
bowtie2-build ./SILVA_138.2_Ref_NR99.fasta ./bowtie2/rRNA

Part 1. Quality Control and Preprocessing

Step 1.1: Quality Control with Fastp

• Purpose: To detect adapter sequences, remove duplicates, filter low-quality sequences, and perform other quality control measures.

  fastp --detect_adapter_for_pe \
        --dedup \
        --dup_calc_accuracy 3 \
        --dont_eval_duplication \
        --qualified_quality_phred 20 \
        --n_base_limit 5 \
        --average_qual 20 \
        --length_required 50 \
        --low_complexity_filter \
        --correction \
        --thread 8 \
        -i ${fq1} \
        -o ${seqID}_r1.fastp.fq.gz \
        -I ${fq2} \
        -O ${seqID}_r2.fastp.fq.gz \
        --json ${seqID}.json \
        --html ${seqID}.html

Step 1.2: Remove rRNA with Bowtie2

• Purpose: To align the cleaned reads against the rRNA database and remove rRNA sequences.

  bowtie2 --local --threads 8 -1 ${seqID}_r1.fastp.fq.gz -2 ${seqID}_r2.fastp.fq.gz -x ./bowtie2/rRNA -S ${seqID}.rRNA.sam --un-conc-gz ${seqID}
  mv ${seqID}.1 ${seqID}.cleanreads.1.fq.gz
  mv ${seqID}.2 ${seqID}.cleanreads.2.fq.gz
  rm -f ${seqID}.rRNA.sam

Part 2. Assembly

Step 2.1: Assembly with Megahit

• Purpose: To assemble the rRNA-removed reads into contigs.

  megahit --memory 20000000000 --min-contig-len 300 -t 12 --out-dir ./megahit --out-prefix ${seqID} -1 ${seqID}.cleanreads.1.fq.gz -2 ${seqID}.cleanreads.2.fq.gz
  perl -pe 's/^>/>${seqID}-/' ./megahit/${seqID}.contigs.fa > ./megahit/${seqID}_addname.fna

Part 3. Identification of RDRP Sequences

Step 3.1: Scan for RDRP with Palmscan

• Purpose: To identify the RNA-dependent RNA polymerase (RDRP) sequences, which are indicative of viral genomes.

palmscan=../bin/palmscan2
getorf -sequence ./megahit/${seqID}_addname.fna -outseq ./megahit/${seqID}_addname.faa -minsize 600
mkdir palmscan_results
${palmscan} -search_pssms ./megahit/${seqID}_addname.faa \
    -tsv palmscan_results/${seqID}.tsv \
    -fev palmscan_results/${seqID}.fev \
    -fasta palmscan_results/${seqID}.pp.fasta \
    -core palmscan_results/${seqID}.core.fasta \
    -report_pssms palmscan_results/${seqID}.report.txt

Part 4. BLASTp against VirusHost Protein Database

Step 4.1: BLASTp for Functional Annotation

• Purpose: To perform a BLASTp search of the assembled contigs against the VirusHost protein database to identify potential viral proteins.

diamond blastp -q palmscan_results/${seqID}.core.fasta -d ./db_diamond/virushostdb_protein.dmnd -o blastp_results.txt --evalue 1e-5 --top 5

• Purpose:Get classification information of virushostdb (./scripts/virushost_db_tax.sh)

#!/bin/bash

#gunzip virushostdb.formatted.cds.faa.gz
input_file="virushostdb.formatted.cds.faa"

awk '
BEGIN { FS="[|]"; OFS="\t" }
/^>/ {
    split($1, id, " ")
    gsub(">", "", id[1])
    print id[1], $4
}
' $input_file >virushostdb.formatted.cds_tax.txt

• Purpose:Get classification information of blastp results

python ./scripts/blastp_tax.py -tax ../virushostdb.formatted.cds_tax.txt -i ./blastp_results.txt  -o ./blastp_results_tax.txt

Script Explanation

The integrated script is in scripts/virus_indentification.sh.

Input File Format

The input file (sample.txt) should contain three columns separated by spaces or tabs. Each row represents a sample with the following fields:

1. fq1: Path to the first FASTQ file (forward reads).
2. fq2: Path to the second FASTQ file (reverse reads).
3. seqID: Sample identifier.

Example of sample.txt

path/to/sample1_R1.fastq.gz /path/to/sample1_R2.fastq.gz sample1
/path/to/sample2_R1.fastq.gz /path/to/sample2_R2.fastq.gz sample2
/path/to/sample3_R1.fastq.gz /path/to/sample3_R2.fastq.gz sample3

Comments: leave a blank line at the end.

Output File Format

scripts/
├── sample1.sh
├── sample2.sh
└── sample3.sh

1 directory, 3 files