---
license: mit

datasets:
  - damlab/HIV_FLT
metrics:
  - accuracy

widget:
 - text: 'C T R P N N N T R K S I R I Q R G P G R A F V T I G K I G N M R Q A H C'
   example_title: 'V3'
 - text: 'M E P V D P R L E P W K H P G S Q P K T A C T N C Y C K K C C F H C Q V C F I T K A L G I S Y G R K K R R Q R R R A H Q N S Q T H Q A S L S K Q P T S Q P R G D P T G P K E S K K K V E R E T E T D P F D' 
   example_title: 'Tat'
 - text: 'P Q I T L W Q R P L V T I K I G G Q L K E A L L D T G A D D T V L E E M N L P G R W K P K M I G G I G G F I K V R Q Y D Q I L I E I C G H K A I G T V L V G P T P V N I I G R N L L T Q I G C T L N F'
   example_title: 'PR'

---

# HIV_BERT model

## Table of Contents
- [Summary](#model-summary)
- [Model Description](#model-description)
- [Intended Uses & Limitations](#intended-uses-&-limitations)
- [How to Use](#how-to-use)
- [Training Data](#training-data)
- [Training Procedure](#training-procedure)
  - [Preprocessing](#preprocessing)
  - [Training](#training)
- [Evaluation Results](#evaluation-results)
- [BibTeX Entry and Citation Info](#bibtex-entry-and-citation-info)

## Summary

The HIV-BERT model was trained as a refinement of the [ProtBert-BFD model](https://huggingface.co/Rostlab/prot_bert_bfd) for HIV centric tasks. It was refined with whole viral genomes from the [Los Alamos HIV Sequence Database](https://www.hiv.lanl.gov/content/sequence/HIV/mainpage.html). This pretraining is important for HIV related tasks as the original BFD database contains few viral proteins making it sub-optimal when used as the basis for transfer learning tasks. This model and other related HIV prediction tasks have been published (link).

## Model Description

Like the original [ProtBert-BFD model](https://huggingface.co/Rostlab/prot_bert_bfd), this model encodes each amino acid as an individual token. This model was trained using Masked Language Modeling: a process in which a random set of tokens are masked with the model trained on their prediction. This model was trained using the damlab/hiv-flt dataset with 256 amino acid chunks and a 15% mask rate.




## Intended Uses & Limitations

As a masked language model this tool can be used to predict expected mutations using a masking approach. This could be used to identify highly mutated sequences, sequencing artifacts, or other contexts. As a BERT model, this tool can also be used as the base for transfer learning. This pretrained model could be used as the base when developing HIV-specific classification tasks.

## How to use

As this is a BERT-style Masked Language learner, it can be used to determine the most likely amino acid at a masked position.

```python
from transformers import pipeline

unmasker = pipeline("fill-mask", model="damlab/HIV_FLT")

unmasker(f"C T R P N [MASK] N T R K S I R I Q R G P G R A F V T I G K I G N M R Q A H C")

[
  {
    "score": 0.9581968188285828,
    "token": 17,
    "token_str": "N",
    "sequence": "C T R P N N N T R K S I R I Q R G P G R A F V T I G K I G N M R Q A H C"
  },
  {
    "score": 0.022986575961112976,
    "token": 12,
    "token_str": "K",
    "sequence": "C T R P N K N T R K S I R I Q R G P G R A F V T I G K I G N M R Q A H C"
  },
  {
    "score": 0.003997281193733215,
    "token": 14,
    "token_str": "D",
    "sequence": "C T R P N D N T R K S I R I Q R G P G R A F V T I G K I G N M R Q A H C"
  },
  {
    "score": 0.003636382520198822,
    "token": 15,
    "token_str": "T",
    "sequence": "C T R P N T N T R K S I R I Q R G P G R A F V T I G K I G N M R Q A H C"
  },
  {
    "score": 0.002701344434171915,
    "token": 10,
    "token_str": "S",
    "sequence": "C T R P N S N T R K S I R I Q R G P G R A F V T I G K I G N M R Q A H C"
  }
]

```

## Training Data

The dataset [damlab/HIV_FLT](https://huggingface.co/datasets/damlab/HIV_FLT) was used to refine the original [rostlab/Prot-bert-bfd](https://huggingface.co/Rostlab/prot_bert_bfd). This dataset contains 1790 full HIV genomes from across the globe. When translated, these genomes contain approximately 3.9 million amino-acid tokens.

## Training Procedure

### Preprocessing

As with the [rostlab/Prot-bert-bfd](https://huggingface.co/Rostlab/prot_bert_bfd) model, the rare amino acids U, Z, O, and B were converted to X and spaces were added between each amino acid. All strings were concatenated and chunked into 256 token chunks for training. A random 20% of chunks were held for validation.

### Training

Training was performed with the HuggingFace training module using the MaskedLM data loader with a 15% masking rate. The learning rate was set at E-5, 50K warm-up steps, and a cosine_with_restarts learning rate schedule and continued until 3 consecutive epochs did not improve the loss on the held-out dataset.


## BibTeX Entry and Citation Info

[More Information Needed]