rbh-specs

These documents are the specifications for the rbh file formats used by the odp software.

The rbh file formats encode orthologies of genomic coordinates. There are two versions of rbh file currently in circulation. The version:

• rbh is the original format that, for each sample for which an orthology is tracked, contains information about the ortholog's scaffold, position, and gene name.
• rbh2 is the new format that encodes the same information as above, but also tracks the strand, and the start/stop position of the feature rather than a generic position coordinate.

General description

The rbh file format was designed with flexibility in mind. In a typical bioinformatics file format, the contents of each column are rigidly specified by their index, and the number of columns in the file must match the specifications. The rbh file format, instead, uses named column headers to determine the column contents. There is a minimum set of columns which must exist for a rbh file, and additional columns can be added at the user's discretion based on their needs. The specifications for the original rbh and the new rbh2 are below.

File Format Specifications - rbh

See the odp documents.

File Format Specifications - rbh2

Every rbh2 file must have the following columns. We recommend that these columns are printed to the file first.

• rbh - contains a unique identifier for this file. Used to trace back to this line from downstream analyses. Do not leave this field blank.
• gene_group - Usually used to mark an ALG.
• color - Used to plot this line. Format is a hex color, like "#B07DF4". If there is no color, use black, "#000000", or nothing, "".

Each sample that is present in the file has five columns. Sample names cannot have underscore characters, _, as this character is used to delimit the fields. Each samplename must be unique. In the example below, we use the string "sampleName" to denote one sample --- this string will be changed for every sample encoded in the genome.

• sampleName_scaf - Contains the scaffold name (fasta header) on which this locus exists. This must be an exact match of a scaffold name fasta header found in the source genome fasta file.
• sampleName_gene - If a protein, this is the fasta header of this protein fasta entry for this ortholog, and must exactly match a protein name in the fasta header. If not a protein, must have some other name unique to this column for this orthology file.
• sampleName_strand - Must be '+' or '-' if the strand of the feature is known. Must be the '.' character if the strand is not known/relevant. If a protein, must be the encoding strand relative to the genome fasta file. If extracted from a .maf file, takes the strand from the alignment.
• sampleName_start - The left-most position of this feature. Uses bed-format numbering conventions. From Wikipedia, this column 'chromStart' is defined as: "Start coordinate on the chromosome or scaffold for the sequence considered (the first base on the chromosome is numbered 0 i.e. the number is zero-based)". This value will always be less than the value of sampleName_stop.
• sampleName_stop - The right-most position of this feature. Uses bed-format numbering conventions. From Wikipedia, this column 'chromEnd' is defined as: "End coordinate on the chromosome or scaffold for the sequence considered. This position is non-inclusive, unlike chromStart (the first base on the chromosome is numbered 1 i.e. the number is one-based)." This value will always be greater than the value of sampleName_start.