hybkit Hyb File Specification

Version: v0.3.4

The ".hyb" (hyb file) format is described by [Travis2014] along with the Hyb software package as a "gff-related format that contains sequence identifiers, read sequences, 1-based mapping coordinates, and annotation information for each chimera".

Each line in a hyb file (a hyb "record") contains information about an RNA sequence read identified as a chimera by an RNA hybridization analysis. Each line contains 15 or 16 columns separated by tab characters ("\t") and provides information on each of the two aligned segments identified within the sequence read. The columns are described as follows by [Travis2014]:

Column 1, unique sequence identifier.

Column 2, read sequence [...].

Column 3, predicted binding energy in kcal/mol.

Columns 4–9, mapping information for first fragment of read: name of matched transcript, coordinates in read, coordinates in transcript, mapping score.

Columns 10–15, mapping information for second fragment of read.

Column 16 (optional, [...]), list of annotations in the format: ‘‘feature1=value1; feature2=value2;..."

The hybkit project uses an extended version of this description, including assigning columns reference names, and defining allowed flags.

Columns

#	Name	Description
1	id	Hybrid Read Identifier
2	seq	Read Nucleotide Sequence
3	energy	Predicted Gibbs Free-Energy of Intra-Hybrid Folding
4	seg1_ref_name	Segment 1 Mapping Reference Identity
5	seg1_read_start	Segment 1 Mapping Start on Read
6	seg1_read_end	Segment 1 Mapping End on Read
7	seg1_ref_start	Segment 1 Mapping Start on Reference
8	seg1_ref_end	Segment 1 Mapping End on Reference
9	seg1_score	Segment 1 Mapping Score
10	seg2_ref_name	Segment 2 Mapping Reference Identity
11	seg2_read_start	Segment 2 Mapping Start on Read
12	seg2_read_end	Segment 2 Mapping End on Read
13	seg2_ref_start	Segment 2 Mapping Start on Reference
14	seg2_ref_end	Segment 2 Mapping End on Reference
15	seg2_score	Segment 2 Mapping Score
16	flags	Hybrid Read Analysis Details

Flags

Hyb Flags:

The following four flags are used by the Hyb software package ([Travis2014]). The definitions provided describe how these flags are used in the hybkit package.

count_total - Integer: Total represented hybrid records, if records have been combined.

count_last_clustering - Integer: Total represented hybrid records at last clustering.

two_way_merged - {"0" or "1"} Boolean representation of whether entries with mirrored 5' and 3' hybrids were merged if the record is a combined record.

seq_IDs_in_cluster - String: Comma-separated list of all record IDs of hybrids merged into this hybrid entry.

hybkit Flags:

The following flags are used by hybkit.

read_count - Integer: Number of sequence reads represented by this record. If the record is combined, this represents the total read count for all merged entries.

orient - String: Orientation of strand. Options: "F" (Forward), "IF" (Inferred Forward), "R" (Reverse), "IR" (Inferred Reverse), "U" (Unknown), or "IC" (Inferred Conflicting).

seg1_type - String: Assigned segment type of segment 1, ex: "miRNA" or "mRNA".

seg2_type - String: Assigned segment type of segment 2, ex: "miRNA" or "mRNA".

seg1_det - String: Arbitrary detail about segment 1.

seg2_det - String: Arbitrary detail about segment 2.

miRNA_seg - String: Indicates which (if any) segment mapping is a miRNA. Options are "N" (none), "5p" (seg1), "3p" (seg2), "B" (both), or "U" (unknown).

target_reg - String: Assigned region of the miRNA target. Options are "5pUTR", "C" ([C]oding), "3pUTR", "NON" ([NON]coding), "N" ([N]one), or "U" ([U]nknown).

ext - Integer: "0" or "1", Boolean representation of whether record sequences were bioinformatically extended as is performed by the Hyb software package.

dataset - String: Label for sequence dataset id (eg. source file), when combining records from different datasets.

Other Details

Item	Role
"\t" (tab)	Column Delimiter
"."	Missing Data Placeholder (equivalent to None)
".hyb"	File Suffix
".hyb.gz"	gzipped File Suffix
Disallowed	Header Lines
Disallowed	In-file Comments

Example

An example .hyb format line (courtesy of [Gay2018]):

2407_718    ATCACATTGCCAGGGATTTCCAATCCCCAACAATGTGAAAACGGCTGTC       .       MIMAT0000078_MirBase_miR-23a_microRNA   1       21      1       21      0.0027  ENSG00000188229_ENST00000340384_TUBB2C_mRNA     23      49      1181    1207    1.2e-06