rpf_Bam2bw

rpf_Bam2bw converts genome-aligned BAM/SAM reads into P-site density tracks in bedGraph or WIG format.

Function

Use this command to generate genome-browser-compatible signal tracks from genome alignment files. The command reads a BAM/SAM file, assigns each read to a P-site according to a P-site offset table, separates plus and minus strand signals, optionally normalizes density to RPM, and writes bedGraph or WIG output.

Input

Required input files:

1. A coordinate-sorted genome alignment file in BAM or SAM format.
2. A P-site offset table generated by RiboParser offset analysis.

The offset table should contain read-length-specific P-site offsets. The current parser expects columns such as:

ribo  length  p_site

Only read lengths present in the offset table are used for P-site assignment.

Parameters

Parameter	Required	Meaning
-b	yes	Input genome alignment file in BAM or SAM format.
-o	yes	Output prefix. If omitted, output naming may be incomplete in the current implementation.
-p	no	P-site offset file in TXT/TSV format.
-f	no	Output format: bedgraph or wig. Default: bedgraph.
-n	no	Normalize read density to RPM. Default: disabled.
-m	no	Merge plus and minus strand density into one output file. Default: disabled.
-t	no	Maximum allowed NH tag value for multi-mapped reads. Reads with NH greater than this value are discarded. Default: 3.
--second	no	Discard secondary alignments. Default: disabled.
--supply	no	Discard supplementary alignments. The flag name is --supply in the current command-line interface. Default: disabled.

Output

When plus and minus strands are kept separate, the command writes:

<prefix>_plus.bedgraph
<prefix>_minus.bedgraph

or, when -f wig is used:

<prefix>_plus.wig
<prefix>_minus.wig

When -m is used, plus and minus strand records are merged into one file:

<prefix>.bedgraph

or:

<prefix>.wig

For bedGraph output, the columns are written without a header:

chrom  start  end  density

Minus-strand density values are written as negative values when strands are kept separate.

Examples

Create strand-specific bedGraph files:

rpf_Bam2bw \
  -b sample1.Aligned.sortedByCoord.out.bam \
  -p sample1_RSBM_offset.txt \
  -f bedgraph \
  -o sample1

Create RPM-normalized bedGraph files and keep only uniquely mapped reads:

rpf_Bam2bw \
  -b sample1.Aligned.sortedByCoord.out.bam \
  -p sample1_RSBM_offset.txt \
  -t 1 \
  --second \
  --supply \
  -f bedgraph \
  -n \
  -o sample1.unique.rpm

Merge plus and minus strands into one bedGraph file:

rpf_Bam2bw \
  -b sample1.Aligned.sortedByCoord.out.bam \
  -p sample1_RSBM_offset.txt \
  -m \
  -n \
  -o sample1.merged.rpm

Write WIG output:

rpf_Bam2bw \
  -b sample1.Aligned.sortedByCoord.out.bam \
  -p sample1_RSBM_offset.txt \
  -f wig \
  -o sample1

Notes

• Use a P-site offset file that matches the same sample type and read-length range used for the BAM/SAM file.
• Use -t 1 when only uniquely mapped reads should be retained according to the NH tag.
• The command writes bedGraph/WIG-style signal tracks. Convert bedGraph to bigWig with external tools such as bedGraphToBigWig if a .bw file is required.
• For genome browser display, make sure chromosome names match the genome assembly used by the browser.