---
license: apache-2.0
inference: false
---

# Enformer

Enformer model. It was introduced in the paper [Effective gene expression prediction from sequence by integrating long-range interactions.](https://www.nature.com/articles/s41592-021-01252-x) by Avsec et al. and first released in [this repository](https://github.com/deepmind/deepmind-research/tree/master/enformer). 

This particular model was trained on sequences of 131,072 basepairs, target length 896 on v3-64 TPUs for 2 and a half days without augmentations and poisson loss. 

This repo contains the weights of the PyTorch implementation by Phil Wang as seen in the [enformer-pytorch repository](https://github.com/lucidrains/enformer-pytorch).

Disclaimer: The team releasing Enformer did not write a model card for this model so this model card has been written by the Hugging Face team.

## Model description

Enformer is a neural network architecture based on the Transformer that led to greatly increased accuracy in predicting gene expression from DNA sequence.

We refer to the [paper](https://www.nature.com/articles/s41592-021-01252-x) published in Nature for details.

### How to use

Refer to the README of [enformer-pytorch](https://github.com/lucidrains/enformer-pytorch) regarding usage.


### Citation info

```
Avsec, Ž., Agarwal, V., Visentin, D. et al. Effective gene expression prediction from sequence by integrating long-range interactions. Nat Methods 18, 1196–1203 (2021). https://doi.org/10.1038/s41592-021-01252-x
```