the following command line should print the distance (5' to 5', in the 1st column) between two mapped reads:

samtools view -f 2 path/to/your.bam |\
awk -F '    ' 'BEGIN  {
      pp=-1;
      pc="";
      }

     {
     if($3!=pc) { pc=$3; pp=int($4);next;}
     if(pp==-1) {pp=int($4);next;}
     printf("%d\t%s\n",int($4)-pp,$0);
     pp=int($4);
     }'

UPDATE: The following C program should print the distance between two mapped reads without any conversion to SAM:

/*
Motivation:
    distance between mapped reads
Author:
    Pierre Lindenbaum PhD
WWW:
    http://plindenbaum.blogspot.com
Contact:
    plindenbaum@yahoo.fr
Compilation:
    gcc -Wall -I $(SAMDIR) -L $(SAMDIR) prog.c -lbam -lz
*/
#include <stdio.h>
#include "bam.h"
#include "sam.h"

static void bamdistance(const char* filename)
 {
 /** previous genomic position in the reference */
 int32_t prev_pos=-1;
 /** previous reference index */
 int32_t prev_reference=-1;
 samfile_t *fp_in = NULL;
 bam1_t *b=NULL;

 fp_in = samopen(filename, "rb", 0);
 if(NULL == fp_in)
      {
      fprintf(stdout,"Could not open file.\n");
      return;
      }
 b = bam_init1();
 while(samread(fp_in, b) > 0) /* loop over the records */
    {
    if(b->core.tid<0 || b->core.pos<0 || (b->core.flag & BAM_FPROPER_PAIR)==0)
            {
            //ignore
            }
       else if(b->core.tid != prev_reference)
        {
        prev_reference=b->core.tid;
        prev_pos=b->core.pos;
        }
        else
            {
            printf("%d\n",(b->core.pos - prev_pos));
            prev_pos=b->core.pos;
            }
      bam_destroy1(b);
      b = bam_init1();
      }
 bam_destroy1(b);
 samclose(fp_in);
 }

int main(int argc, char *args[])
  {
  int optind=1;
  if(optind==argc)
      {
      fprintf(stderr,"Illegal number of arguments\n");
      return EXIT_FAILURE;
      }
  /* loop over the files */
  while(optind

You can use SAMTOOLS. Using the command line:

samtools view MyFile.bam

will allow you to read your BAM file as a SAM without decompressing. Then, you can pipe this command and use grep/awk or whatever. For example:

samtools view -X MyFile.bam | less -S

I always use Bio::DB::Sam modules to access bam files. Although the name is 'sam', they provide very flexible functions to access almost everything from the bam files, too. The downside is that you have to know a bit perl programming to do your queries, but it's definitely worthwhile to learn. Plus, the manual page is very well-written and with a lot useful examples: http://search.cpan.org/~lds/Bio-SamTools/lib/Bio/DB/Sam.pm

It does help to know what you want it for. And of course perl and Bio::DB::Sam are certainly irreplaceable. However, you could also try samtools view (and maybe tview) option to output a bam file on to the screen (and pipe to a file). There seems to be an option for specifying Chromosome, start and end values as well but I haven't tried that part. http://samtools.sourceforge.net/samtools.shtml

Hello Sir,

I would like to view a SRR038263.sra BAM file in samtools and do alignment to reference genome. How could I view the SRR038263.sra in samtools? I have already install and make samtools at same directory where I have SRR038263.sra BAM file. I would be glad for your kindness and support.

Regards, Rocky

The format of BAM is documented in the same document as SAM, currently http://samtools.sourceforge.net/SAM1.pdf. Without your being more specific, the only answer I can give is "Write a binary parser for BAM. It's not that friggin' hard."

I may have more detailed answers if you specify what information you look for, what language you're using, or what it is you didn't understand about the BAM documentation.