What Does This Tool Do?
3DPredictor web tool can predict 3D genome contacts map for a chromosome region using information on CTCF binding and gene expression (RNA-Seq data).
Files can be uploaded from local storage or external public resources, e.g. ENCODE, by FTP. 3DPredictor is definitely omnivorous thing, so it can be fed with gzipped and bzipped files too.
Input data must be narrowPeak format. No header, tab-separated table only is allowed.
chr1 840081 840400 treat1_peak_1 69 . 4.89872 10.50944 6.91052 158 chr1 919419 919785 treat1_peak_2 87 . 5.85158 12.44148 8.70936 130 chr1 937220 937483 treat1_peak_3 66 . 4.87632 10.06728 6.61759 154
Columns significant for prediction are the following:
RNA-Seq data is to contain fields gene_id and FPKM. gene_id can be Ensemble ID or gene name. This table must have header.
gene_id Gene Name Reference Strand Start End Coverage FPKM TPM ENSG00000185960.8 SHOX chrX + 624344 646823 0.000000 0.000000 0.000000 ENSG00000002834.13 SHOX chrX + 624344 659411 0.000000 0.000000 0.000000
Genome assembly should be relevant to your CTCF ChIP-seq data. For RNA-seq data, genome version has no importance, because genomic coordinates are obtained from Ensembl for each gene/transcript. 3DPredictor does not use coordinates provided in the uploaded RNA-Seq data.
Chromosomal region for prediction must correspond to the format you can see on placeholder. Chromosome name is to start with chr, and commas are to be used as digit group separators.
Trained models for human or mouse data at 5kb resolution are provided. We strongly recommend to use model which was not trained on the chromosome of interest, to avoid overfitting. E.g., if you need prediction for chr1, please use a model trained on even chromosomes.
3DPredictor produces a gzipped HiC map with 3D-genome contacts. Further it's sent to your e-mail with brief prediction report. You can visualize this file using Juicebox. If your job failed, please contact email@example.com.
Enjoy the prediction! We hope it might be useful for your research ^_^
Polina Belokopytova, Evgeniy Mozheiko, Miroslav Nuriddinov, Daniil Fishman, Veniamin Fishman. Quantitative prediction of enhancer-promoter interactions
Recent experimental and computational efforts provided large datasets describing 3-dimensional organization of mouse and human genomes and showed interconnection between expression profile, epigenetic status and spatial interactions of loci. These interconnections were utilized to infer spatial organization of chromatin, including enhancer-promoter contacts, from 1-dimensional epigenetic marks. Here we showed that predictive power of some of these algorithms is overestimated due to peculiar properties of biological data. We proposed an alternative approach, which gives high-quality predictions of chromatin interactions using only information about gene expression and CTCF-binding. Using multiple metrics, we confirmed that our algorithm could efficiently predict 3-dimensional architecture of normal and rearranged genomes.
Page Development & Design
Polina Belokopytova, Emil Valeev