Documentation And Resource Guide
Help
This page is the documentation hub for score interpretation, search entry points, identifier support, release content, and contact information for collaboration or data questions.
Scores FAQ
How To Interpret Scores
Does a high score mean a true interaction probability?
No. P2PSigLip scores are prioritisation scores primarily intended to rank candidate pairs within a query or candidate pool. They are not calibrated interaction probabilities and are not equivalent to experimental validation.
How should I use the ranking in practice?
Use the ranking to shortlist candidates for structure-aware modelling, hypothesis generation, and targeted validation such as Y2H, Co-IP, BiFC, or related assays.
Can I apply one universal score threshold across datasets?
No. The score is best read as a prioritisation signal within a query or candidate pool, not as a universal yes/no decision boundary that transfers cleanly across datasets or biological contexts.
What other evidence should be read together with the score?
Combine the score with annotations, homolog support, local network context, prior literature, and downstream experiments before making strong biological claims.
Search FAQ
Finding Genes And Proteins
What is the fastest way to start?
Start from one species on the homepage, open the analysis workspace for that dataset, and search by gene ID or representative protein ID.
Can I search with aliases and protein IDs?
Yes. Gene IDs, representative protein IDs, and accepted aliases are resolved through the dataset-specific ID mapping table whenever mappings are available.
What becomes available after opening one result page?
You can inspect the canonical gene, representative protein, aliases, sequence, annotations, top-ranked partners, homologs, enrichment entry points, and local network tools.
Data FAQ
Release Content And Scope
What data are included in one released dataset?
Each dataset can include ranked prediction rows, canonical IDs, representative proteins, alias mappings, representative sequences, annotations, and linked downstream analysis views.
Why do some genes have richer annotations than others?
Annotation depth depends on the underlying species resources and mapping coverage. Some datasets are richer in ontology terms, aliases, or homolog support than others.
Where should I go for the manuscript summary and citation?
Visit the Paper page for the manuscript-oriented summary, citation, and figure-inspired highlights.
Contact FAQ
Research And Resource Contacts
For collaboration, data questions, or additional proteome-scale prediction requests, use the contacts below.
Zhang Zhao Laboratory · China Agricultural University
Room 332, Horticulture Building, China Agricultural University, Beijing, China
Corresponding Lab Contact
Prof. Zhao Zhang
For scientific contact, collaboration, and institutional correspondence.
Site Maintainer
JiCheng Tang
For website issues, content corrections, and implementation questions.
Code repository: GitHub