update figure

Author: etowahadams

NOTICE

@@ -1,5 +1,5 @@
 foldsteer
-Copyright 2026 the foldsteer authors
+Copyright 2026 the authors
 
 Licensed under the Apache License, Version 2.0. See LICENSE.
 

README.md

@@ -1,36 +1,36 @@
 # foldsteer
 
-Inference-time chemical steering for AF3-style biomolecular structure
-prediction. Improves the chemical validity of predicted ligand geometry
-(stereochemistry, planarity, bond geometry, steric clash) without retraining
-and without modifying the host model's weights.
-
-Implements the two mechanisms used by Boltz-1x and Protenix-v2:
+Implementes basic chemical steering for OpenFold3-p2. Reimplements methods
+previously developed in Boltz-1x and Protenix-v2:
 
 * **physical guidance** — gradient descent on a flat-bottom chemical energy,
   applied to the denoiser's `x0` prediction at each diffusion step;
 * **Feynman-Kac steering** — sample a particle population and resample it
   toward low-energy trajectories.
 
+The package is set up in a model agnostic way. Adaptors can be written for new models. 
+ 
 ## Benchmark on OF3p2
 
 OpenFold3-p2 (155k) on the 100 smallest Runs N' Poses post-2025 targets, 1 seed
 × 5 samples, guidance only (20 GD steps), scored with PXMeter. Rates are over
-the 103 ligand chains, each represented by its **top-1 sample ranked by
-`chain_pair_iptm`**.
+the 103 ligand chains, each represented by its top-1 sample ranked by
+`chain_pair_iptm`. Ligand success = RMSD < 2 Å and lDDT-PLI > 0.8.
+
+![benchmark](benchmark.png)
 
 | | baseline | steered |
 | --- | --- | --- |
-| PoseBusters valid | 66.0% | **89.3%** |
-| PB valid & ligand RMSD < 2 Å & lDDT-PLI > 0.8 | 38.8% | **47.6%** |
-| ligand RMSD < 2 Å & lDDT-PLI > 0.8 | 49.5% | 50.5% |
-| wall clock, 4× GH200 | 9m56s | 17m11s (**1.73×**) |
+| PoseBusters valid | 66.0% | 89.3% |
+| ligand success | 49.5% | 50.5% |
+| PB valid & ligand success | 38.8% | 47.6% |
+| wall clock, 4× GH200 | 9m56s | 17m11s (1.73×) |
 
-Steering fixes chemistry without moving docking accuracy: paired ligand by
-ligand it fixed 25 validity failures and broke 1 (McNemar p < 1e-4), while
-accuracy moved by a net +1 (p = 1.0). Nearly all of the gain is sterics —
-`minimum_distance_to_protein` failures fall 141 → 3 across all 515 poses;
-chirality is second (31 → 13).
+Steering fixes chemistry without moving placement: paired ligand by ligand it
+fixed 25 validity failures and broke 1 (McNemar p < 1e-4), while ligand success
+moved by a net +1 (p = 1.0). Nearly all of the gain is sterics —
+`minimum_distance_to_protein` failures fall 25 → 0; chirality is second
+(9 → 4).
 
 Caveat: 10 of 161 multi-atom ligand chains were skipped because RDKit rejected
 the molecule rebuilt from `atom_array` (valence errors on quaternary nitrogen
@@ -132,3 +132,8 @@ OF3 inference input — which turned up three silent failures in it (integer
 `tests/test_of3_extraction.py`. Guidance is now benchmarked against real OF3
 inference (see Benchmark above); Feynman-Kac steering is not, and the formal
 charge gap in the `AtomArray` rebuild is the clearest outstanding fix.
+
+## Process 
+
+Using Claude Science, I wrote a delibrated about what the API should look like. 
+Then Claude code implemented the plan.