bHLHDB: A next generation database of basic helix loop helix transcription factors based on deep learning model

Area of Science:

• Plant molecular biology
• Bioinformatics
• Genomics

Background:

• The basic helix loop helix (bHLH) superfamily is crucial for plant development and metabolism.
• bHLH transcription factors regulate gene expression through dimerization.
• Increasing molecular data necessitates organized storage and analysis.

Purpose of the Study:

• To develop and implement a comprehensive relational database for all plant bHLH superfamily members.
• To create a user-friendly platform for querying bHLH family and sequence information.
• To integrate advanced analytical tools for efficient bHLH research.

Main Methods:

• Compilation of all known plant bHLH superfamily members.
• Development of a relational database (bHLHDB) with query functionalities.
• Integration of Hidden Markov Model (HMM) and BLAST search.
• Implementation of a deep learning model for transcription factor (TF) type prediction.

Main Results:

• Creation of bHLHDB (www.bhlhdb.org), a centralized resource for plant bHLH factors.
• Successful integration of HMM and BLAST for sequence analysis.
• Development of a deep learning model achieving 97.54% accuracy and 97.76% precision in TF classification.
• The database is publicly accessible to the scientific community.

Conclusions:

• bHLHDB is a unique, next-generation database for bHLH transcription factors.
• The integrated deep learning model offers rapid and accurate TF identification.
• This database is expected to be an invaluable tool for future bHLH family studies.