对不起,此内容只适用于EN

(tentative plan, to be updated)

 

AIPC (Artificial Intelligence Proteomics Competition) series focuses on leveraging AI to uncover the ‘dark matter’ of MS-based proteomics.

 

1. Competition task

DDA rescoring optimization: Participants will develop AI models to refine peptide-spectrum matching and enhance the ranking of potential peptide sequences using machine learning techniques and existing protein databases.

 

2. Participants

Students, researchers, bioinformaticians, computational biologists, AI practitioners, and anyone worldwide interested in applying AI to proteomics.

 

3. Sponsors

 

 

4. MSDT Dataset

Each row in the MSDT files represents a peptide-spectrum matches (PSM).

We plan to launch two competition tracks: the Enthusiast Track and the Professional Track, each corresponding to training datasets of different scales.

The Enthusiast Track will include approximately 16 million PSMs, while the Professional Track will consist of around 300 million PSMs.

The test set consists of approximately 3 million spectra with corresponding annotations and is shared between both competition tracks.

 

5. Baseline Model

We will offer a DDA-rescoring baseline to help participants know how to use MSDT files and train a base model on GPUs.

 

6. Evaluation Metrics

  • Identification Quantity: Under a two-species scenario, with conservative FDR < 1% and decoy FDR < 1%, the number of PSMs correctly identified by the model.
  • Evaluation Time: The model must complete processing a single file within 3 hours; otherwise, it will not be evaluated.

 

7. Competition Rules

7.1 Team Formation:

Each team must have a minimum of one and a maximum of five members.

Code sharing between teams is strictly prohibited—violators will be disqualified.

7.2 Submission Rules:

Each team can submit up to two times per day, with a total submission limit of 100 within 100 days.

Submissions exceeding six hours of runtime will be considered invalid.

Invalid submissions will not count toward the total submission limit.

7.3 Ranking Rules:

A/B Leaderboard System:

  • The test dataset is split into A leaderboard (40%) and B leaderboard (60%).
  • The A leaderboard updates in real-time and displays rankings.
  • The B leaderboard (used to determine the final awards) will be revealed three days after the competition ends.

Final Ranking Criteria:

  • Score > Submission Count > Submission Time
  • If two teams have the same Score, the team with fewer submissions ranks higher.
  • If both Score and submission count are identical, the team that submitted earlier ranks higher.

Final Submission Selection:

  • Each team’s leader can designate two final submissions for the B leaderboard ranking.
  • If no selection is made, the highest-ranked A leaderboard submission will be used by default.

 

8. Anticipated Competition Duration

May – Aug 2025  (100 days)

 

9. Awards & Prizes

First Prize: $5,000 (each track)

Second Prize: $1,500 (each track)

Third Prize: $500 (each track)

 

10. Additional Benefits

Computing Credit Rewards: Every registered participant will receive a ¥100 Bohrium® computing credit.

Best Notebook Award: Participants must submit complete code using the Bohrium Notebook platform. We encourage contestants to publish relevant content in Notebook format on the Case Plaza with the tag AI4SCUP-[AIPC]. The top three notebooks with the most likes will each receive a ¥1,000 computing credit.

Internship Opportunities: Outstanding participants may be recommended for internship opportunities at relevant institutions, gaining access to top-tier research resources and networking with leading interdisciplinary experts.

Invited tour in participating institutes.