README.md

---
tags:
- dna
- human_genome
---

# GENA-LM (gena-lm-bert-base-t2t)

GENA-LM is a Family of Open-Source Foundational Models for Long DNA Sequences.

GENA-LM models are transformer masked language models trained on human DNA sequence.

Differences between GENA-LM (`gena-lm-bert-base-t2t`) and DNABERT:
- BPE tokenization instead of k-mers;
- input sequence size is about 4500 nucleotides (512 BPE tokens) compared to 512 nucleotides of DNABERT
- pre-training on T2T vs. GRCh38.p13 human genome assembly.

Source code and data: https://github.com/AIRI-Institute/GENA_LM

Paper: https://www.biorxiv.org/content/10.1101/2023.06.12.544594

This repository also contains models that are finetuned on downstream tasks:
- promoters predictions (branch [promoters_300_run_1](https://huggingface.co/AIRI-Institute/gena-lm-bert-base-t2t/tree/promoters_300_run_1))
- splice sites prediction (branch [spliceai_run_1](https://huggingface.co/AIRI-Institute/gena-lm-bert-base-t2t/tree/spliceai_run_1))
- epigenetic features and gene expression (trained on enformer dataset, branch [enformer](https://huggingface.co/AIRI-Institute/gena-lm-bert-base-t2t/tree/enformer))

and models that are used in our [GENA-Web](https://dnalm.airi.net) web tool for genomic sequence annotation:
- deepsea (gena_web_deepsea, branch [gena_web_deepsea](https://huggingface.co/AIRI-Institute/gena-lm-bert-base-t2t/tree/gena_web_deepsea))
- deepstarr (gena_web_deepstarr, branch [gena_web_deepstarr](https://huggingface.co/AIRI-Institute/gena-lm-bert-base-t2t/tree/gena_web_deepstarr))
 

## Examples

### How to load pre-trained model for Masked Language Modeling
```python
from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained('AIRI-Institute/gena-lm-bert-base-t2t')
model = AutoModel.from_pretrained('AIRI-Institute/gena-lm-bert-base-t2t', trust_remote_code=True)

```

### How to load pre-trained model to fine-tune it on classification task
Get model class from GENA-LM repository:
```bash
git clone https://github.com/AIRI-Institute/GENA_LM.git
```

```python
from GENA_LM.src.gena_lm.modeling_bert import BertForSequenceClassification
from transformers import AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained('AIRI-Institute/gena-lm-bert-base-t2t')
model = BertForSequenceClassification.from_pretrained('AIRI-Institute/gena-lm-bert-base-t2t')
```
or you can just download [modeling_bert.py](https://github.com/AIRI-Institute/GENA_LM/tree/main/src/gena_lm) and put it close to your code.

OR you can get model class from HuggingFace AutoModel:
```python
from transformers import AutoTokenizer, AutoModel
model = AutoModel.from_pretrained('AIRI-Institute/gena-lm-bert-base-t2t', trust_remote_code=True)
gena_module_name = model.__class__.__module__
print(gena_module_name)
import importlib
# available class names:
# - BertModel, BertForPreTraining, BertForMaskedLM, BertForNextSentencePrediction,
# - BertForSequenceClassification, BertForMultipleChoice, BertForTokenClassification,
# - BertForQuestionAnswering
# check https://huggingface.co/docs/transformers/model_doc/bert
cls = getattr(importlib.import_module(gena_module_name), 'BertForSequenceClassification')
print(cls)
model = cls.from_pretrained('AIRI-Institute/gena-lm-bert-base-t2t', num_labels=2)
```

## Model description
GENA-LM (`gena-lm-bert-base-t2t`) model is trained in a masked language model (MLM) fashion, following the methods proposed in the BigBird paper by masking 15% of tokens. Model config for `gena-lm-bert-base-t2t` is similar to the bert-base:

- 512 Maximum sequence length
- 12 Layers, 12 Attention heads
- 768 Hidden size
- 32k Vocabulary size

We pre-trained `gena-lm-bert-base-t2t` using the latest T2T human genome assembly (https://www.ncbi.nlm.nih.gov/assembly/GCA_009914755.3/). The data was augmented by sampling mutations from 1000-genome SNPs (gnomAD dataset). Pre-training was performed for 2,100,000 iterations with batch size 256 and sequence length was equal to 512 tokens. We modified Transformer with [Pre-Layer normalization](https://arxiv.org/abs/2002.04745), but without the final layer LayerNorm.

## Evaluation
For evaluation results, see our paper: https://www.biorxiv.org/content/10.1101/2023.06.12.544594v1


## Citation
```bibtex
@article{GENA_LM,
	author = {Veniamin Fishman and Yuri Kuratov and Maxim Petrov and Aleksei Shmelev and Denis Shepelin and Nikolay Chekanov and Olga Kardymon and Mikhail Burtsev},
	title = {GENA-LM: A Family of Open-Source Foundational Models for Long DNA Sequences},
	elocation-id = {2023.06.12.544594},
	year = {2023},
	doi = {10.1101/2023.06.12.544594},
	publisher = {Cold Spring Harbor Laboratory},
	URL = {https://www.biorxiv.org/content/early/2023/06/13/2023.06.12.544594},
	eprint = {https://www.biorxiv.org/content/early/2023/06/13/2023.06.12.544594.full.pdf},
	journal = {bioRxiv}
}
```