Help

General information

What is ZRdb?

ZRdb is a database that serves to describe both KZFPs and repeats (including LTR, SINE, LINE and SVAs). The database contains information on the sequences, domains and DNA binding specificities of various KZFPs, while also providing information such as the classification and consensus sequences of different repeats. By utilising the existing ChIP-seq data of KZFPs, ZRdb also describes the binding repression and co-evolution relationship between KZFPs and different repeats, ranking these relationships according to reliability.

Where does the ZRdb data come from?

The descriptive data of KZFPs in ZRdb — including annotation score, sequences, cellular localisation, molecular function and involved cellular pathway information — were sourced from annotations on UniProt. For the DNA binding specificity of each KZFP, we searched for its ChIP-seq data on Cistrome. Where ChIP-seq data was available, we used the motif information calculated by MEME-ChIP to describe its DNA binding specificity; where it was unavailable, we used the predicted motif from the Zinc Finger Recognition Code as a replacement.

The annotated data about repeats in ZRdb primarily comes from Dfam, including the description, classification and consensus sequences of each repeat. To describe the distribution of various repeats on the human genome, we used the repeat annotation files for hg19 and hg38 provided in IGV.

What is the binding-probability rating of KZFPs for retrotransposons?

To enhance the reliability of our calculations and provide a multidimensional evaluation of KZFP binding to retrotransposons, we developed a rating system. It categorises the likelihood of a KZFP binding a retrotransposon into three levels, from low to high:

Level 1 — peak overlap

ChIP-seq peaks that intersect with specific repeat sequences in the genome, calculated from Cistrome bed files and human repeat annotation.

Level 2 — motif in repeat

The most confident motif position from MEME-ChIP analysis is located within the retrotransposon sequence to calculate binding probability.

Level 3 — motif in consensus

Builds on level 2 by locating the corresponding region of the retrotransposon type on its consensus sequence.

Glossary

Domains of KZFPs

KRAB, ZF and degenerate ZF domains of KZFPs.

Mya / Myr

Million years ago / million years.

Time of emergence

The time when a repeat or KZFP first appears in the genome.

Time of change in binding specificity

The time at which a KZFP undergoes a change in DNA-binding specificity, resulting in a new specificity identical to that of humans.

Time correlation

The proximity of the emergence of a repeat and a KZFP. Strength is categorised into three levels: highly correlated (emergence times precisely match), moderately correlated (the repeat emerged earlier than the KZFP by no more than 15 Myr), and weakly correlated (the repeat emerged later than the KZFP, or earlier by more than 15 Myr).

Gene Ontology annotation of KZFPs

The Gene Ontology describes knowledge of the biological domain through three aspects: molecular function, biological process and cellular component.

Is there ChIP-seq data

Whether ChIP-seq data for the KZFP is available in the Cistrome database.

Consensus sequence of repeats

Consensus sequences obtained from Dfam, derived from the seed alignment using a specialised caller.

Repeat structure

For certain families there is a curated collection of TE-derived proteins forming a non-redundant set; several families may therefore lack directly linked coding sequences. Annotation for these families is obtained from Dfam.

Motif (MEME-ChIP)

DNA-binding motifs of KZFPs obtained using MEME-ChIP and mapped with pLogo.

Motif (predicted)

DNA-binding motifs predicted using the Zinc Finger Recognition Code.

Peak overlapping

A ChIP-seq peak of a KZFP that intersects with a certain repeat sequence in the genome.

Motif in repeats

The motif of a KZFP identified by MEME-ChIP is present within repeat sequences.

Motif in repeats consensus

The motif of a KZFP identified by MEME-ChIP can be located at the corresponding region of the repeat consensus sequence.