Training Notebook: A guide to fine-tuning the ProkBERT-mini model for promoter identification tasks.
Evaluation Notebook: Demonstrates how to evaluate the finetuned ProkBERT-mini-promoter model on test datasets.

Model Application

The model was trained for binary classification to distinguish between promoter and non-promoter sequences. The length and composition of the promoter sequences were standardized to ensure compatibility with alternative methods and to facilitate direct comparison of model performance.

Simple Usage Example

The following example demonstrates how to use the ProkBERT-mini-promoter model for processing a DNA sequence:

python

from prokbert.prokbert_tokenizer import ProkBERTTokenizer
from prokbert.models import BertForBinaryClassificationWithPooling
finetuned_model = "neuralbioinfo/prokbert-mini-promoter"
kmer = 6
shift= 1

tok_params = {'kmer' : kmer,
             'shift' : shift}
tokenizer = ProkBERTTokenizer(tokenization_params=tok_params)
model = BertForBinaryClassificationWithPooling.from_pretrained(finetuned_model)
sequence = 'TAGCGCATAATGATTTCCTTATAAGCGATCGCTCTGAAAGCGTTCTACGATAATAATGATATCCTTTCAATAATAGCGTAT'
inputs = tokenizer(sequence, return_tensors="pt")
# Ensure that inputs have a batch dimension
inputs = {key: value.unsqueeze(0) for key, value in inputs.items()}
# Generate outputs from the model
outputs = model(**inputs)
print(outputs)

Model Details

Developed by: Neural Bioinformatics Research Group

Architecture:

Traditionally, models like ...SequenceClassification classify sequences based on the hidden representation of the [CLS] or starting token. However, in our approach, we utilize the base model enhanced with a pooling layer that integrates information across all nucleotides in the sequence. The input is expected to be 80bp long, same as in the dataset.

Tokenizer: The model uses a 6-mer tokenizer with a shift of 1 (k6s1), specifically designed to handle DNA sequences efficiently.

Parameters:

Parameter	Description
Model Size	20.6 million parameters
Max. Context Size	1024 bp
Training Data	206.65 billion nucleotides
Layers	6
Attention Heads	6

Intended Use

Intended Use Cases: ProkBERT-mini-k6-s1 is intended for bioinformatics researchers and practitioners focusing on genomic sequence analysis, including:

sequence classification tasks
Exploration of genomic patterns and features

Installation of ProkBERT (if needed)

For setting up ProkBERT in your environment, you can install it using the following command (if not already installed):

python

try:
    import prokbert
    print("ProkBERT is already installed.")
except ImportError:
    !pip install prokbert
    print("Installed ProkBERT.")

Training Data and Process

Overview: The model was pretrained on a comprehensive dataset of genomic sequences to ensure broad coverage and robust learning.

Masking performance of the ProkBERT family.

Evaluation of Promoter Prediction Tools on E-coli Sigma70 Dataset

Tool	Accuracy	MCC	Sensitivity	Specificity
ProkBERT-mini	0.87	0.74	0.90	0.85
ProkBERT-mini-c	0.87	0.73	0.88	0.85
ProkBERT-mini-long	0.87	0.74	0.89	0.85
CNNProm	0.72	0.50	0.95	0.51
iPro70-FMWin	0.76	0.53	0.84	0.69
70ProPred	0.74	0.51	0.90	0.60
iPromoter-2L	0.64	0.37	0.94	0.37
Multiply	0.50	0.05	0.81	0.23
bTSSfinder	0.46	-0.07	0.48	0.45
BPROM	0.56	0.10	0.20	0.87
IBPP	0.50	-0.03	0.26	0.71
Promotech	0.71	0.43	0.49	0.90
Sigma70Pred	0.66	0.42	0.95	0.41
iPromoter-BnCNN	0.55	0.27	0.99	0.18
MULTiPly	0.54	0.19	0.92	0.22

The ProkBERT family models exhibit remarkably consistent performance across the metrics assessed. With respect to accuracy, all three tools achieve an impressive

Metric	ProkBERT-mini	ProkBERT-mini-c	ProkBERT-mini-long	Promotech	Sigma70Pred	iPromoter-BnCNN	MULTiPly
Accuracy	0.81	0.79	0.81	0.61	0.62	0.61	0.58
F1	0.81	0.78	0.81	0.43	0.58	0.65	0.58
MCC	0.63	0.57	0.62	0.29	0.24	0.21	0.16
Sensitivity	0.81	0.75	0.79	0.29	0.52	0.66	0.57
Specificity	0.82	0.82	0.83	0.93	0.71	0.55	0.59

Promoter prediction performance metrics on a diverse test set. A comparative analysis of various promoter prediction tools, showcasing their performance across key metrics including accuracy, F1 score, MCC, sensitivity, and specificity.

Ethical Considerations and Limitations

As with all models in the bioinformatics domain, ProkBERT-mini-promoter should be used responsibly. Testing and evaluation have been conducted within specific genomic contexts, and the model's outputs in other scenarios are not guaranteed. Users should exercise caution and perform additional testing as necessary for their specific use cases.

Reporting Issues

Please report any issues with the model or its outputs to the Neural Bioinformatics Research Group through the following means:

Model issues: GitHub repository link
Feedback and inquiries: obalasz@gmail.com

Reference

If you use ProkBERT-mini in your research, please cite the following paper:

@ARTICLE{10.3389/fmicb.2023.1331233,
    AUTHOR={Ligeti, Balázs and Szepesi-Nagy, István and Bodnár, Babett and Ligeti-Nagy, Noémi and Juhász, János},
    TITLE={ProkBERT family: genomic language models for microbiome applications},
    JOURNAL={Frontiers in Microbiology},
    VOLUME={14},
    YEAR={2024},
    URL={https://www.frontiersin.org/articles/10.3389/fmicb.2023.1331233},
    DOI={10.3389/fmicb.2023.1331233},
    ISSN={1664-302X},
    ABSTRACT={...}
}