Important
Ribbit is under active development. Please refer to Change log and make sure to get the latest version of the repository.
Ribbit identifies tandem repeat (TR) regions of variable motif sizes in DNA sequences. Tandem repeats are DNA segments consisting of two or more nearly identical copies of a motif occurring contiguously. Ribbit is designed to improve motif decomposition and resolve complex repeat structures, accurately detecting TRs with motif sizes up to 100 bp.
The algorithm converts DNA sequences to 2-bit format and uses basic bit operations to deliniate potential repetitive stretches of a certain periodicity. The DNA sequence of a potential repetitive stretch is decomposed to identify a representative motif of identified peridicity. There are two different approaches to identifying the representative motif depending the on the peridicity. The sequence in the potential stretch is aligned with the perfect repeat of the identified representative motif to calculate the purity of the repeat. The potential sequence is either trimmed or dropped based on the user desired purity and minimum repeat length thresholds. A given sequence is idependently processed for potential repeats of all user-desired motif sizes. The results from each idependent search are merged and compared to resolve for nested/overlapping interpretations of a sequence as repetitive sequence of different periodicities. Overlapping tandem repeat interpretations of different motif sizes are retained/dropped based on preference for more pure stretches.
The conversion of DNA to 2-bit stretches results in fast identification of potential repetitive stretches and allows the time for careful motif decomposition of repeat sequence. Ribbit provides a comprehensive bed file as an output and takes about 5-7 secs to resolve an MB of DNA sequence. The program can also be run on multi-threaded mode making it ideal for processing large genomes.
To compile Ribbit, please follow these instructions:
sudo apt-get install boost
sudo apt-get install zlib1g-devgit clone https://github.com/SowpatiLab/ribbit.git
git checkout dev
cd ribbit
make$ ./ribbit [options] -i sequence.fasta
# with output file provided
$ ./ribbit [options] -i sequence.fasta -o results.bed
# pip input from standard input
$ cat sequence.fasta | ./ribbit [options] -i -
$ echo "ATgcatgcGGAGGAGGAGGAGGAGGAcagtcgata" | ./ribbit -i - ./ribbit -h
Ribbit: identification of tandem repeats and annotation of complex TRs in genomes
Version: 1.0.2
Options for running the tool:
-h [ --help ] Ribbit detects tandem repeat regions in DNA, accurately resolving
complex repeat structures and motif sizes up to 100 bp.
--version Prints out the version of ribbit.
-i [ --input-file ] arg Input sequence. Can be a fasta file (optionally gzipped) or '-' for
stdin.
-o [ --output-file ] arg File path for output file. Default: stdout
-m [ --min-motif-length ] arg The minimum length of the motif of the TR loci. [int] Default: 2
-M [ --max-motif-length ] arg The maximum length of the motif of the TR loci. [int] Default: 100
-p [ --min-purity ] arg The minimum allowed purity of repeat sequence. Purity is calculated
as the (matches/(matches+mismatches+indels)) in the alignment of
region sequence to perfect repeat of consensus motif. [float]
Default: 0.8
-q [ --min-motif-purity ] arg Minimum purity of each motif with consensus motif. Calculated as
the average of (matches/(matches+mismatches+indels)) for each motif
length in the alignment of region sequence to perfect repeat of
consensus motif. [float] Default: 0.8
-l [ --min-length ] arg The minimum length of the repeat. Input can be an integer or a
tab-separated file with two columns of motif length and the length
cutoff. [int or file] Default: 12 for STRs (motif length <= 6),
2*(motif length) for others.
--min-units arg The minimum number of units of the repeat. Input can be a integer
or a tab-separated file with two columns, first is the motif size
and second unit cutoff. [int or file] Default: 2 for all motif
sizes.
--perfect-units arg The minimum number of complete units with 100% match with the
consensus motif in the repeat. Input can be an integer or a
tab-separated file with two columns of the motif length and the
unit cutoff. [int or file] Default: 2
--cigar Include cigar string of the alignment of the sequence with the
perfect repeat of the consensus motif in the output. Default:
false.
-t [ --threads ] arg Number of threads to be used for running. [int] Default: 1After identifying a potential repetitive region with a given periodicity, Ribbit determines the consensus motif
of the tandem repeat. The sequence is then aligned to a perfect tandem repeat generated from this consensus motif. Repeat purity
is calculated as the number of matching bases divided by the alignment length between the sequence and the perfect repeat. The
--min-purity option allows users to set the minimum purity threshold for tandem repeat detection. It accepts a floating-point
value between 0 and 1.
$purity = matches / (matches + mismatches + indels)$
Motif purity is purity calculated for each motif stretch and is then averaged across all the motifs. --min-motif-purity option
allows users to set the minimum threshold for a TR. This is used to trim the motifs of lesser purity from the edges of the repeat and
report the stretch with average motif purity greater than or equal to the user defined threshold. It accepts a floating-point value
between 0 and 1.
motif purity at the ith motif is calculated as
$motif\_purity_i = matches / ((matches + mismatches + indels))$
motif purity is calculated as the average across all motif units
$motif\_purity = average(motif\_purity_0 + motif\_purity_1 +... + motif\_purity_n)$
| S.No | Column | Description |
|---|---|---|
| 1 | Chromosome | chromosome or sequence name as specified in fasta header |
| 2 | Repeat Start | 0-based start position of tandem repeat in the chromosome |
| 3 | Repeat Stop | end position of tandem repeat in the chromosome |
| 4 | Motif | sequence of unit motif of tandem repeat |
| 5 | Purity | purity of the repeat |
| 6 | Motif length | length of the unit motif in bases |
| 7 | Repeat Length | length of the repeat based on the coordinates |
| 8 | Repeat Units | number of unit motifs of the TR |
| 9 | Info | additional information of the repeat region including the CIGAR string of alignment and information of the nested TRs in the region. |
| chromosome | start | end | motif | purity | motif_length | repeat_length | repeat_units | info |
|---|---|---|---|---|---|---|---|---|
| test | 292 | 305 | AC | 0.93 | 2 | 13 | 6 | I |
| test | 481 | 496 | AGC | 0.82 | 3 | 15 | 5 | I |
| test | 827 | 843 | GCCCAGGT | 1.00 | 8 | 16 | 2 | I |
| test | 1017 | 1032 | TGCGGAG | 0.93 | 7 | 15 | 2 | I |
| test | 1508 | 1523 | AGGC | 0.81 | 4 | 15 | 3 | I |
| test | 1682 | 1833 | TGGAGGGTGGGGCCAAATGGAAGTGGGCGGGGCTGTGG | 0.90 | 38 | 151 | 3 | I |
| test | 1863 | 1890 | AGGGC | 0.83 | 5 | 27 | 5 | M:1869-1890-6-0.81:AGGGGC |
| test | 2182 | 2197 | CCGGT | 0.93 | 5 | 15 | 3 | I |
| test | 2277 | 2296 | TGGCCTCC | 1.00 | 8 | 19 | 2 | I |
The INFO field provides detailed information about tandem repeat (TR) structure, purity, and nested sub-repeats.
Each attribute in the field is separated by a colon (:).
<M_or_I> : <subrepeat_info> : <motifs>if the --cigar option is enabled:
<M_or_I> : <CIGAR> : <subrepeat_info> : <motifs> : <subrepeat_CIGARs>| Attribute | Description |
|---|---|
| Type | Indicates whether the repeat is isolated (I) or contains nested, more pure repeats (M). |
| CIGAR (optional) | If the --cigar option is enables this reports the CIGAR string of the alignment of sequence with a perfect repeat of the consensus motif. |
| 2nd – Subrepeat info | Lists subrepeats in the format {start}-{stop}-{period}-{purity}, separated by commas. Example: 1068-1115-6-1.00 means a subrepeat from position 1068–1115, with motif length 6 bp and 100% purity. |
| 3rd – Motifs | Provides the motifs of the corresponding subrepeats, separated by commas. |
| Subrepeat CIGAR strings (optional) | If the --cigar option is used, lists the CIGAR strings of each subrepeat, separated by commas. |
M:1068-1115-6-1.00,1103-1126-11-1.00,1114-1126-6-1.00:AACCCT,accctaaccct
#with --cigar enabled
M100M:1068-1115-6-1.00,1103-1126-11-1.00:AACCCT,accctaaccct:47M,23MIf you found ribbit useful, we would appreciate it if you could cite our manuscript: Ribbit: Accurate identification and annotation of complex tandem repeat sequences in genomes
- Removed bug to exclude N at the end of the repeat.
- Optimized alignment for seeds larger than 10kb.
- Fixed concatenating outputs in multi-threaded mode.
- Taking sequence input from standard input.
- Default output directed to standard output.
- Fixed seed sorting for seed with identical coordinates.
- Ribbit underwent major changes in this version. The algorithmic logic for identifying the potential tandem repeats (seed TRs) has changed.
- The output format has been changed. TR regions are reported with the nested nested repeats reported in the
infocolumn.
Anukrati Sharma
Akshay Kumar Avvaru
For queries or suggestions, please contact:
Akshay Kumar Avvaru - avvaruakshay@gmail.com
Divya Tej Sowpati - tej@ccmb.res.in