Step-by-step guide for BAM/SAM to FASTA conversion

Here is a step-by-step guide for BAM/SAM to FASTA conversion, tailored to beginners and incorporating the latest tools and techniques:

Introduction

Converting BAM/SAM files to FASTA format can be achieved efficiently using modern bioinformatics tools such as samtools and seqtk, among others. This manual outlines the methods step-by-step and includes the use of Unix commands and optional Perl scripting for advanced customization.

Prerequisites

1. Tools Required:
   • samtools (Version ≥ 1.3 recommended)
   • seqtk (if additional processing is needed)
   • Optional: Perl interpreter for custom scripts
2. Installation:
   • Install samtools:
     bash

     Copy code

     sudo apt-get install samtools

   • Install seqtk:
     bash

     Copy code

     git clone https://github.com/lh3/seqtk.git
cd seqtk
make
sudo cp seqtk /usr/local/bin/

3. Input Files:
   • A BAM/SAM file (input.bam or input.sam)
   • Reference FASTA file (if required for strand-specific processing)

Method 1: Using samtools fasta Command

Steps:

1. Convert BAM to FASTA:
   bash

   Copy code

   samtools fasta input.bam > output.fasta

   • This method extracts the sequence information directly.
2. For strand-specific data, use the -F or -f flags to filter specific flags. For example:
   bash

   Copy code

   samtools view -F 16 input.bam | samtools fasta - > output.fasta

   • Here, -F 16 excludes reverse-strand reads.

Method 2: Using samtools and seqtk

Steps:

1. Convert BAM to FASTQ using samtools bam2fq:
   bash

   Copy code

   samtools bam2fq input.bam | seqtk seq -A > output.fasta

   • The -A flag in seqtk converts FASTQ to FASTA.
2. For strand-specific processing:
   bash

   Copy code

   samtools view -F 16 input.bam | samtools bam2fq - | seqtk seq -A > output.fasta


Method 3: Using BBMap’s reformat.sh

Steps:

1. Download and install BBMap:
   bash

   Copy code

   wget https://sourceforge.net/projects/bbmap/files/latest/download
unzip BBMap*.zip
cd bbmap

2. Convert BAM to FASTA:
   bash

   Copy code

   ./reformat.sh in=input.bam out=output.fasta


Method 4: Using Perl Script for Custom BAM/SAM to FASTA Conversion

Steps:

1. Create a Perl script (bam_to_fasta.pl):
   perl

   Copy code

   #!/usr/bin/perl
   use strict;
   use warnings;

   while (<STDIN>) {
chomp;
my @fields = split(/\t/, $_);
if ($fields[0] !~ /^@/) { # Skip header lines
my ($read_name, $sequence) = ($fields[0], $fields[9]);
print ">$read_name\n$sequence\n";
}
}


2. Save the script and make it executable:
   bash

   Copy code

   chmod +x bam_to_fasta.pl

3. Run the script:
   bash

   Copy code

   samtools view input.bam | ./bam_to_fasta.pl > output.fasta


Method 5: Using BEDOPS Toolkit

Steps:

1. Install BEDOPS:
   bash

   Copy code

   sudo apt-get install bedops

2. Convert BAM to BED and then to FASTA:
   bash

   Copy code

   bam2bed < input.bam | bed2faidxsta.pl > output.fasta


Tips for Strand-Specific Sequencing

• When dealing with strand-specific sequencing data, ensure the correct strand is chosen:
  • -F 16 excludes reverse-strand reads.
  • -f 16 includes only reverse-strand reads.

Best Practices

1. Always index the BAM file before conversion:
   bash

   Copy code

   samtools index input.bam

2. Use compressed output for large datasets:
   bash

   Copy code

   samtools fasta input.bam | gzip > output.fasta.gz


Troubleshooting

1. samtools version mismatch:
   • Update to the latest version of samtools:
     bash

     Copy code

     conda install -c bioconda samtools

2. Missing seqtk:
   • Install seqtk from its GitHub repository as shown above.
3. Custom FASTA header:
   • Modify the Perl script to include additional information in the FASTA header if needed.

This manual ensures beginner-friendly, step-by-step instructions for converting BAM/SAM files to FASTA format. Select the method that best fits your data and computational requirements.