Allows users to search MAAD entries using:

• Antibody/Nanobody Name (e.g., "PDI-106", "MxR")
• Target Virus (e.g., SARS-CoV-1, Influenza)
• Germline V/J Gene (e.g., IGHV3-30)
• Developmental Origin (e.g., Human, Mouse)
• Project ID (e.g., PubMed ID: 37075126, Patent ID: CN100374464)

Clicking on a search result leads to a detailed entry page with all associated metadata, while clicking on a PDB number or project number (such as PMID or patent ID) directs users to the original source on the respective PDB website or database.

Step 1: Click the right-hand slider to choose a search type

Step 2: Type keywords

The search will match sequences of identical length, leveraging N-Gram indexing and LIKE queries to identify similar or embedded subsequences:

• Input: Provide individual CDR sequences (CDR1/2/3) and select a numbering scheme (IMGT, Kabat, or Chothia) and chain type (Heavy or Light).
• Output:
  • Matched entries filtered by CDR length or sequence similarity.
  • Interactive visualizations:
    • Sequence logo plots (via WebLogo) constructed for CDR regions of identical length.
    
    • Pairwise alignment visualization with mismatch highlighting for equal-length sequences.

MAAD performs BLAST-based sequence alignment, querying both antibody nucleotide and amino acid sequences against the database:

• Input: Submit a variable region (VH/VL) nucleic acid or amino acid sequence.
• Process: MAAD performs a BLAST search against the database.
• Output:
  • A ranked list of the top 200 matched entries by sequence similarity.
  • Interactive visualizations:
    • CDR and V/J germline genes of the input sequence.
    
    • V/J gene usage dot plot (germline distribution).
    
    • Pairwise alignment viewer with mismatch highlighting.

MAAD provides antigen-antibody complex structures with detailed interface annotations and allows direct search by PDB ID on the search page.

• Step 1: Click the right-hand slider to choose a search type.

• Step 2: View your results.

Users can input their own heavy chain nucleotide sequences and select target viral antibody sequences for joint analysis. Results can be downloaded from the results page. Tree construction may take some time, usually less than 3 minutes.

The Download module allows users to export customized or full datasets from MAAD in standardized .xlsx format. For each entry in MAAD, the following data is available for download:

• Sequence Data:
  • Full-length amino acid (VH/VL) and corresponding nucleotide sequences (if available).
  • GenBank accession IDs (for nucleotide sequences).
• CDR Annotation:
  • Annotated CDR1/2/3 regions of heavy and light chains using three schemes: IMGT, Kabat, and Chothia.
  • CDR lengths under each numbering convention.
• Functional Information:
  • Confirmed activity against (binding and/or neutralization) target strain.
  • Target antigens (e.g., RBD, HA, F protein).
• Structural Metadata:
  • Resolved structure IDs and direct hyperlinks to PDB entries.
• Reference Information:
  • Source publication (PubMed ID or patent number).
  • Curation date (inclusion and publication timestamp).

Provides an overview of the MAAD dataset, enabling users to explore key trends and distributions:

• Total number and distribution of antibodies/nanobodies per target antigen (Antibody Distribution).
• Usage frequency of V/J genes across different antigens (Genotype Distribution). Genotype frequencies are categorized as: Low: all genotypes ranked within the top 30 by frequency; Medium: the top 50 most prevalent genotypes; High: the top 100 most prevalent genotypes.
• CDR1/2/3 length distribution (CDR Statistics).

For questions, feature requests, or reporting issues, please contact us at:

wangjy@ibp.ac.cn

liyx@ibp.ac.cn