Biopanning data bank 2018: hugging next generation phage display

Bifang He, Lixu Jiang, Yaocong Duan, Guoshi Chai, Yewei Fang, Juanjuan Kang, Min Yu, Ning Li, Zhongjie Tang, Peng‐Cheng Yao, Pengcheng Wu, Ratmir Derda, Jian HuangDatabase 2018 · DOI

TL;DR

The Biopanning Data Bank (BDB) has been significantly upgraded to incorporate next-generation sequencing (NGS) data from phage display experiments, alongside traditional Sanger sequencing data. This update addresses the growing need for a comprehensive resource to analyze the vast datasets generated by modern high-throughput screening methods, aiming to facilitate the identification of specific peptide ligands for various applications.

What Problem Does This Paper Address?

Phage display is a powerful technique for identifying peptides that bind to specific target molecules, ranging from proteins and nucleic acids to cells and tissues. The traditional method, however, relies on Sanger sequencing, which is low-throughput and only provides a glimpse into the vast sequence space of the displayed peptides. This limitation, coupled with the prevalence of "target-unrelated peptides" (TUPs) that can dominate screening results, hinders the efficient identification of truly specific binders. The advent of next-generation sequencing (NGS) has revolutionized phage display by enabling the analysis of millions of sequences, but this also creates a new challenge: managing and interpreting these massive datasets. Existing databases and analytical tools often struggle with the scale and complexity of NGS data, particularly in distinguishing genuine binders from TUPs.

Methods

The core contribution of this paper is the significant upgrade and expansion of the Biopanning Data Bank (BDB). Key methodological advancements include:

  • Inclusion of Next-Generation Phage Display (NGPD) Data: The BDB now incorporates data from NGPD experiments, which utilize high-throughput sequencing technologies. This includes data from both selected populations and unselected libraries (naïve libraries), providing a crucial baseline for comparison.
  • New Data Fields: To accommodate the richer information from NGS, new data fields have been added, such as sequencing platform and sequencing method. Curation information, including the curator and curation date, has also been enhanced.
  • Improved Browsing and Searching: The BDB's user interface has been redesigned for better user experience. Users can now browse traditional and NGPD data separately. The advanced search system has been expanded to include more data fields, offering prompts and hints for more efficient data retrieval.
  • Integration of Standalone Tools: The BDB now integrates standalone tools for analyzing biopanning data, including those for identifying TUPs. These tools are available in both graphical user interface (GUI) and command-line versions, allowing for local analysis of both traditional and NGPD data.
  • Enhanced 3D Structure Viewer: The interactive 3D chemical structure viewer has been re-implemented using JSmol, a web-based viewer that supports devices without Java installed, improving accessibility and functionality.

Datasets

The BDB database itself serves as the primary dataset. The 2018 update includes:

  • 3291 sets of biopanning data collected from 1540 published articles.
  • This comprises 95 NGPD data sets and 3196 traditional biopanning data sets.
  • Data is categorized into nine target types, with proteins, cells, and inorganic molecules being the most common.
  • The database also contains 29,795 peptides, 1971 targets, 545 templates, 484 combinatorial libraries, and 318 three-dimensional structures.

Key Insights and Ideas

  • The necessity of NGPD data integration: The paper highlights that traditional phage display methods are becoming insufficient due to the limitations of Sanger sequencing and the challenge of TUPs. Integrating NGPD data is crucial for a comprehensive understanding of phage display results.
  • The challenge of TUPs persists with NGPD: While NGPD offers high-throughput analysis, the problem of identifying target-unrelated peptides remains. The inclusion of standalone tools within BDB aims to address this by providing users with methods to filter out non-specific binders.
  • User experience is paramount: The redesign of the BDB homepage and the enhancement of the search system demonstrate a focus on making the vast amount of data accessible and usable for researchers.
  • BDB as a central hub: The paper positions BDB as a vital resource for the biopanning community, enabling cross-checking of results and facilitating ligand identification studies.

Key Results

  • The BDB database has been significantly expanded, now housing 3291 biopanning data sets, a substantial increase from previous versions.
  • The inclusion of 95 NGPD data sets marks a significant step towards accommodating modern high-throughput screening data.
  • The integration of standalone TUP analysis tools provides users with practical utilities for data interpretation.
  • The improved 3D structure viewer enhances the visualization of target-template and target-peptide complexes.

Limitations and Future Work

  • Limitations: The paper notes that the "on-the-fly data visualization tool, sequences features, is disabled when a set of phage display data contains <3 or >200 peptides." This suggests a limitation in the real-time analysis capabilities for very small or very large datasets within the browsing interface. Additionally, while the standalone tools can handle NGPD data, the paper implies that some older web-based tools might still be limited to small-scale data.
  • Future Work: The authors plan to continue maintaining and updating the BDB database. They also intend to integrate more computational tools for clustering analysis, motif analysis, and epitope mapping. A significant future development will be the construction of an experimentally supported dataset, which will further enhance the database's utility.

Summary updated

Abstract

The BDB database is available at http://immunet.cn/bdb.
BibTeX
@inproceedings{he2018biopanning,
  author = {Bifang He and Lixu Jiang and Yaocong Duan and Guoshi Chai and Yewei Fang and Juanjuan Kang and Min Yu and Ning Li and Zhongjie Tang and Peng‐Cheng Yao and Pengcheng Wu and Ratmir Derda and Jian Huang},
  title = {{Biopanning data bank 2018: hugging next generation phage display}},
  booktitle = {Database},
  year = {2018},
  doi = {10.1093/database/bay032},
}
Figure