TABLE 1.
Tools for modeling and predicting chromatin interactions.
| Tool name | Input features | Target features | Method/algorithm |
| See review by Xu et al. (2018) | Histone marks, TFs binding, DHS | Promoter–enhancer interactions | See review by Xu et al. (2018) |
| MacPherson et al. (2018) model | HP1, H3K9me3 | Compartments | Polymer modeling |
| MichroM + MEGABASE (Di Pierro et al.) | Histone marks, TFs binding | Compartments | NN classifier + polymer modeling |
| Huang et al. (2015) model | Histone marks | TADs | BART |
| 3Disease Browser (Li et al., 2016) | Enhancers and TAD boundaries | Rearranged TADs | Linear model |
| Lollipop (Kai et al., 2018) | Chip-seq data, CTCF directionality | Loops | ML ensemble classifier (random forest) |
| 3DEpiloop (Al Bkhetan and Plewczynski, 2018) | Histone marks, TFs binding | Loops | ML ensemble classifier (random forest) |
| CTCF-MP (Zhang et al., 2018) | CTCF binding, DHS, nucleotide sequence | Loops | ML ensemble classifier/NN (Boosted trees/word2vec) |
| EpiTensor (Zhu et al., 2016) | Histone marks, TFs binding | Loops | Tensor modeling + PCA |
| DeepMILO (Trieu et al., 2020) | Sequence of loop anchors | Rearranged loops | CNN and RNN |
| 3D-GNOME (Sadowski et al., 2019) | CTCF ChIA-PET | Rearranged loops | linear models |
| 3DPredictor (Belokopytova et al., 2020) | CTCF, RNA-seq | Whole hi-c map | ML ensemble regression (gradient boosting) |
| Hi-C Reg (Zhang et al., 2019) | Histone marks, TFs binding, DHS | Whole hi-c map | ML ensemble regression (random forest) |
| Akita (Fudenberg et al., 2020) | Sequence | Whole hi-c map | CNN |
| DeepC (Schwessinger et al., 2020) | Sequence | Whole hi-c map | CNN |
| Yifeng Qi and Bin Zhang model (Qi and Zhang, 2019) | CTCF binding, Chromatin states | Whole hi-c map | Polymer modeling |
| HiP-HoP (Buckle et al., 2018) | CTCF and cohesin binding, Histone marks or DHS | Whole hi-c map | Polymer modeling |
| Rowley et al. (2017) model | GRO-seq + CTCF binding | Whole hi-c map | Explicit algebraic model |
| PRISMR (Bianco et al., 2018) | Wild-type Hi-C data | Whole hi-c map in mutated cells | Polymer modeling |
DHS, DNAse I hypersensitivity sites; TFs, transcription factors; TADs, topologically associated domains; ML, machine learning; NN, neural network; CNN, convolutional neural network; RNN, recurrent neural network; BART, Bayesian additive regression trees; PCA, principle component analysis.