Public

Cryo-EM Single Particle Analysis — Public Tier

(5 scenes)

Full-access development tier with all data visible.

What you get

Measurements (y), ideal forward operator (H), spec ranges, ground truth (x_true), and true mismatch spec.

How to use

Load HDF5 → compare reconstruction vs x_true → check consistency → iterate.

What to submit

Reconstructed signals (x_hat) and corrected spec as HDF5.

Parameter Specifications

✓

True spec visible — use these exact values for Scenario III oracle reconstruction.

Parameter Spec Range True Value Unit
defocus_error -100.0 – 200.0 50.0 nm
astigmatism -20.0 – 40.0 10.0 nm
beam_tilt -0.2 – 0.4 0.1 mrad
ice_thickness_variation 44.0 – 62.0 53.0 nm

InverseNet Baseline Scores

Method: CPU_baseline — Mismatch parameter: nominal

Scenario I (Ideal)

5.70 dB

SSIM 0.0146

Scenario II (Mismatch)

4.40 dB

SSIM -0.0052

Scenario III (Oracle)

25.22 dB

SSIM 0.0919

Per-scene breakdown (4 scenes)
Scene PSNR I SSIM I PSNR II SSIM II PSNR III SSIM III
scene_00 5.68 0.0131 4.37 -0.0049 27.06 0.1087
scene_01 5.64 0.0164 4.41 0.0041 23.62 0.0568
scene_02 5.81 0.0181 4.47 -0.0131 22.68 0.0815
scene_03 5.67 0.0107 4.34 -0.0071 27.53 0.1205

Public Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 DiffusionCryo + gradient 0.842 37.04 0.978 0.9 ✓ Certified Luo et al., arXiv 2024
2 CryoSTAR + gradient 0.825 35.48 0.971 0.91 ✓ Certified Yang et al., Nat. Methods 2024
3 CryoFormer + gradient 0.811 34.54 0.965 0.9 ✓ Certified Gao et al., CVPR 2024
4 cryoDRGN + gradient 0.789 32.27 0.946 0.94 ✓ Certified Zhong et al., Nat. Methods 2021
5 CryoGEM + gradient 0.788 33.21 0.954 0.87 ✓ Certified He et al., NeurIPS 2023
6 IsoNet + gradient 0.723 28.48 0.891 0.91 ✓ Certified Liu et al., Nat. Commun. 2021
7 cryoSPARC + gradient 0.667 25.92 0.83 0.88 ✓ Certified Punjani et al., Nat. Methods 2017
8 RELION-3D + gradient 0.640 24.18 0.775 0.94 ✓ Certified Scheres, J. Struct. Biol. 2012
9 CTFFIND4 + gradient 0.556 21.11 0.651 0.92 ✓ Certified Rohou & Grigorieff, J. Struct. Biol. 2015

Visible Data Fields

y H_ideal spec_ranges x_true true_spec

Dataset

Format: HDF5
Scenes: 5

Scoring Formula

0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)

PSNR: 40% SSIM: 40% Consistency: 20%
Back to Cryo-EM Single Particle Analysis