Skip to content

Latest commit

 

History

History
93 lines (79 loc) · 6.53 KB

File metadata and controls

93 lines (79 loc) · 6.53 KB

AlphaFold2 Algorithm Index

This index provides a comprehensive mapping of all 32 algorithms from the AlphaFold2 supplementary materials to their source code implementations and explanation notebooks.

Legend

  • ✅ Complete (includes pseudocode image, source code reference, NumPy implementation, and test examples)

Algorithms

# Algorithm Notebook Source File Status
1 MSA Block Deletion algorithm-1-MSABlockDeletion.ipynb model/tf/data_transforms.py
2 Inference algorithm-2-Inference.ipynb model/modules.py:AlphaFold
3 Input Embedder algorithm-3-InputEmbedder.ipynb model/modules.py:EmbeddingsAndEvoformer
4 relpos algorithm-4-relpos.ipynb model/modules.py:EmbeddingsAndEvoformer
5 one_hot algorithm-5-one_hot.ipynb jax.nn.one_hot
6 Evoformer Stack algorithm-6-EvoformerStack.ipynb model/modules.py:EvoformerIteration
7 MSA Row Attention with Pair Bias algorithm-7-MSARowAttentionWithPairBias.ipynb model/modules.py:MSARowAttentionWithPairBias
8 MSA Column Attention algorithm-8-MSAColumnAttention.ipynb model/modules.py:MSAColumnAttention
9 MSA Transition algorithm-9-MSATransition.ipynb model/modules.py:Transition
10 Outer Product Mean algorithm-10-OuterProductMean.ipynb model/modules.py:OuterProductMean
11 Triangle Multiplication (Outgoing) algorithm-11-TriangleMultiplicationOutgoing.ipynb model/modules.py:TriangleMultiplication
12 Triangle Multiplication (Incoming) algorithm-12-TriangleMultiplicationIncoming.ipynb model/modules.py:TriangleMultiplication
13 Triangle Attention (Starting Node) algorithm-13-TriangleAttentionStartingNode.ipynb model/modules.py:TriangleAttention
14 Triangle Attention (Ending Node) algorithm-14-TriangleAttentionEndingNode.ipynb model/modules.py:TriangleAttention
15 Pair Transition algorithm-15-PairTransition.ipynb model/modules.py:Transition
16 Template Pair Stack algorithm-16-TemplatePairStack.ipynb model/modules.py:TemplatePairStack
17 Template Pointwise Attention algorithm-17-TemplatePointwiseAttention.ipynb model/modules.py:Attention
18 Extra MSA Stack algorithm-18-ExtraMsaStack.ipynb model/modules.py:ExtraMsaStack
19 MSA Column Global Attention algorithm-19-MSAColumnGlobalAttention.ipynb model/modules.py:GlobalAttention
20 Structure Module algorithm-20-StructureModule.ipynb model/folding.py:StructureModule
21 Rigid from 3 Points algorithm-21-rigidFrom3Points.ipynb model/r3.py:rigids_from_3_points
22 Invariant Point Attention algorithm-22-InvariantPointAttention.ipynb model/folding.py:InvariantPointAttention
23 Backbone Update algorithm-23-BackboneUpdate.ipynb model/folding.py:FoldIteration
24 Compute All Atom Coordinates algorithm-24-computeAllAtomCoordinates.ipynb model/all_atom.py
25 makeRotX algorithm-25-makeRotX.ipynb model/all_atom.py
26 Rename Symmetric Ground Truth Atoms algorithm-26-renameSymmetricGroundTruthAtoms.ipynb model/folding.py:compute_renamed_ground_truth
27 Torsion Angle Loss algorithm-27-torsionAngleLoss.ipynb model/folding.py:supervised_chi_loss
28 Compute FAPE algorithm-28-computeFAPE.ipynb model/all_atom.py:frame_aligned_point_error
29 Predict Per-Residue LDDT algorithm-29-predictPerResidueLDDT.ipynb model/modules.py:PredictedLDDTHead
30 Recycling (Inference) algorithm-30-RecyclingInference.ipynb model/modules.py:AlphaFold
31 Recycling (Training) algorithm-31-RecyclingTraining.ipynb model/modules.py:AlphaFold
32 Recycling Embedder algorithm-32-RecyclingEmbedder.ipynb model/modules.py:EmbeddingsAndEvoformer

Categories

Data Preprocessing

  • Algorithm 1: MSA Block Deletion

Embedding

  • Algorithm 3: Input Embedder
  • Algorithm 4: relpos (Relative Position Encoding)
  • Algorithm 5: one_hot

Evoformer

  • Algorithm 6: Evoformer Stack
  • Algorithm 7: MSA Row Attention with Pair Bias
  • Algorithm 8: MSA Column Attention
  • Algorithm 9: MSA Transition
  • Algorithm 10: Outer Product Mean
  • Algorithm 11-12: Triangle Multiplication (Outgoing/Incoming)
  • Algorithm 13-14: Triangle Attention (Starting/Ending Node)
  • Algorithm 15: Pair Transition

Templates

  • Algorithm 16: Template Pair Stack
  • Algorithm 17: Template Pointwise Attention

Extra MSA

  • Algorithm 18: Extra MSA Stack
  • Algorithm 19: MSA Column Global Attention

Structure Module

  • Algorithm 20: Structure Module
  • Algorithm 21: Rigid from 3 Points
  • Algorithm 22: Invariant Point Attention (IPA)
  • Algorithm 23: Backbone Update
  • Algorithm 24: Compute All Atom Coordinates
  • Algorithm 25: makeRotX

Losses

  • Algorithm 26: Rename Symmetric Ground Truth Atoms
  • Algorithm 27: Torsion Angle Loss
  • Algorithm 28: Compute FAPE
  • Algorithm 29: Predict Per-Residue LDDT (pLDDT)

Recycling

  • Algorithm 30: Recycling (Inference)
  • Algorithm 31: Recycling (Training)
  • Algorithm 32: Recycling Embedder

Main Pipeline

  • Algorithm 2: Inference