B

Birna Bert

Developed by buetnlpbio
A Transformer encoder model based on BERT architecture, specifically designed for generating RNA sequence embeddings
Downloads 364
Release Time : 7/2/2024

Model Overview

BiRNA-BERT is a BERT model specialized for processing RNA sequences, capable of generating nucleotide-level and sequence-level embeddings, suitable for bioinformatics applications

Model Features

Dual-mode embedding generation
Supports embedding generation via both BPE token units and single nucleotides, providing representations at different granularities
Sequence length extension
Configurable maximum sequence length extension, default support for 1024, extendable up to 2048
Bioinformatics optimization
Specifically designed for RNA sequence processing, suitable for bioinformatics application scenarios

Model Capabilities

RNA sequence embedding generation
Nucleotide-level feature extraction
Sequence-level feature extraction

Use Cases

Bioinformatics
RNA sequence analysis
Analyzing RNA sequence characteristics through embedding representations
RNA structure prediction
Providing feature inputs for RNA secondary structure prediction
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase