> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/google-deepmind/alphafold3/llms.txt
> Use this file to discover all available pages before exploring further.

# Confidence Metrics

> Understanding pLDDT, PAE, pTM, and ipTM confidence scores

## Overview

AlphaFold 3 provides multiple confidence metrics to assess prediction quality. These metrics help you understand:

* **Local accuracy**: How confident is each atom/residue?
* **Global accuracy**: Is the overall fold correct?
* **Interface quality**: Are chain-chain interactions accurate?
* **Relative positions**: How accurate are token-token distances?

<Note>
  Similar to AlphaFold 2 and AlphaFold-Multimer, but with enhanced metrics for ligands, nucleic acids, and multi-chain complexes.
</Note>

## pLDDT: Per-Atom Confidence

<CardGroup cols={1}>
  <Card title="predicted Local Distance Difference Test" icon="atom">
    **Range**: 0-100 (higher = more confident)

    **Granularity**: Per-atom (more detailed than AlphaFold 2's per-residue)

    **What it measures**: Predicted accuracy of local structure
  </Card>
</CardGroup>

### Understanding pLDDT Values

<Tabs>
  <Tab title="Confidence Ranges">
    | pLDDT Score | Interpretation       | Typical Usage                                        |
    | ----------- | -------------------- | ---------------------------------------------------- |
    | **90-100**  | Very high confidence | Well-ordered regions, reliable for detailed analysis |
    | **70-90**   | Confident            | Generally accurate, suitable for most applications   |
    | **50-70**   | Low confidence       | Use with caution, may be flexible/disordered         |
    | **\< 50**   | Very low confidence  | Likely disordered, unreliable coordinates            |
  </Tab>

  <Tab title="Polymer-Specific">
    ### Proteins

    * Similar to AlphaFold 2's lDDT-Cα metric
    * Per-atom granularity (not just Cα)
    * Considers distances to other polymer atoms within 15Å

    ### DNA/RNA

    * Uses wider 30Å radius (vs. 15Å for proteins)
    * Reflects nucleic acid structural characteristics
    * Modified LDDT considers errors only to polymers

    ### Ligands

    * Modified LDDT considers only ligand-to-polymer distances
    * Does NOT consider ligand-to-ligand atom distances
    * Focuses on protein-ligand interface quality
  </Tab>
</Tabs>

### Accessing pLDDT Values

In the confidences JSON:

```json theme={null}
{
  "atom_plddts": [95.2, 94.8, 93.1, ...],  // num_atoms array
  "atom_chain_ids": ["A", "A", "A", ...]   // maps atoms to chains
}
```

### Visualization

Most structure viewers can color by B-factor. AlphaFold 3 stores pLDDT in the B-factor column:

```python PyMOL theme={null}
# Load and color by pLDDT (stored in B-factor)
load model.cif
spectrum b, blue_white_red, minimum=50, maximum=90
```

## PAE: Predicted Aligned Error

<CardGroup cols={1}>
  <Card title="Positional Uncertainty" icon="arrows-cross">
    **Range**: 0-32 Ångströms (lower = more confident)

    **Granularity**: Per-token pair

    **What it measures**: Error in relative position/orientation between two tokens
  </Card>
</CardGroup>

### How PAE Works

PAE(i, j) estimates the error in token j's position when the structure is aligned using token i's coordinate frame.

<Steps>
  <Step title="Alignment">
    Align predicted structure to true structure using token i's frame
  </Step>

  <Step title="Measurement">
    Measure positional error of token j after alignment
  </Step>

  <Step title="Prediction">
    PAE predicts this error in Ångströms
  </Step>
</Steps>

### Interpreting PAE Plots

<Note>
  PAE is visualized as a 2D heatmap where both axes represent token indices.
</Note>

**Low PAE (dark blue)**: Confident in relative positions

* **Diagonal blocks**: Well-structured domains
* **Off-diagonal blocks**: Confident interfaces/contacts

**High PAE (yellow/red)**: Uncertain relative positions

* Between domains: Flexible linkers
* Between chains: Uncertain/no interaction

### Frame Construction

<Tabs>
  <Tab title="Proteins & Nucleic Acids">
    Frames constructed from backbone atoms:

    * **Proteins**: Same as AlphaFold 2 (Cα, N, C)
    * **Nucleic acids**: Backbone phosphate groups
  </Tab>

  <Tab title="Small Molecules">
    Frames constructed for each atom from:

    * The atom itself
    * Its closest neighbors in reference conformer
    * Enables PAE for ligands and PTMs
  </Tab>
</Tabs>

### Accessing PAE Values

In the confidences JSON:

```json theme={null}
{
  "pae": [[0.5, 1.2, 15.3, ...],    // num_tokens × num_tokens
          [1.3, 0.4, 14.8, ...],
          [15.1, 14.9, 0.6, ...]],
  "token_chain_ids": ["A", "A", "B", ...]  // maps tokens to chains
}
```

### Use Cases

<CardGroup cols={2}>
  <Card title="Domain Organization" icon="puzzle-piece">
    Low within-domain PAE, high between-domain PAE suggests multi-domain protein with flexible linkers
  </Card>

  <Card title="Interaction Confidence" icon="handshake">
    Low off-diagonal PAE between chains indicates confident protein-protein interface
  </Card>

  <Card title="Binding Assessment" icon="magnet">
    High PAE between protein and ligand suggests uncertain binding pose
  </Card>

  <Card title="Relative Orientation" icon="compass">
    PAE reveals which regions have well-defined spatial relationships
  </Card>
</CardGroup>

## pTM: Predicted Template Modeling Score

<CardGroup cols={1}>
  <Card title="Global Structure Confidence" icon="globe">
    **Range**: 0-1 (higher = more confident)

    **Granularity**: Global scalar

    **What it measures**: Accuracy of entire predicted structure
  </Card>
</CardGroup>

### Interpretation Guidelines

<Tabs>
  <Tab title="Overall Quality">
    | pTM Score   | Interpretation                                 |
    | ----------- | ---------------------------------------------- |
    | **> 0.8**   | Very confident, high-quality prediction        |
    | **0.5-0.8** | Decent confidence, likely correct overall fold |
    | **\< 0.5**  | Low confidence, fold may be incorrect          |
  </Tab>

  <Tab title="Size Considerations">
    <Warning>
      pTM is strict for small structures. Values \< 0.05 for structures with fewer than 20 tokens.
    </Warning>

    For small structures or short chains:

    * **Use PAE or pLDDT instead** for quality assessment
    * pTM becomes more reliable with >20 tokens
    * ipTM more informative for multi-chain systems
  </Tab>
</Tabs>

### TM-Score Background

The Template Modeling score measures structural similarity:

$$
\text{TM-score} = \max \left[ \frac{1}{L} \sum_{i=1}^{L} \frac{1}{1 + (d_i / d_0)^2} \right]
$$

Where:

* **L**: Length of target sequence
* **d\_i**: Distance between aligned residues
* **d\_0**: Normalization factor (length-dependent)

<Note>
  References: [Zhang & Skolnick, 2004](https://doi.org/10.1002/prot.20264); [Xu & Zhang, 2010](https://doi.org/10.1093/bioinformatics/btq066)
</Note>

### Accessing pTM

In summary confidences JSON:

```json theme={null}
{
  "ptm": 0.84,
  "chain_ptm": [0.89, 0.76, 0.92]  // per-chain pTM
}
```

## ipTM: Interface Predicted TM Score

<CardGroup cols={1}>
  <Card title="Interface Quality" icon="link">
    **Range**: 0-1 (higher = more confident)

    **Granularity**: Global or per-chain-pair

    **What it measures**: Accuracy of predicted interfaces between chains
  </Card>
</CardGroup>

### Interpretation Guidelines

| ipTM Score  | Interpretation                    | Action                     |
| ----------- | --------------------------------- | -------------------------- |
| **> 0.8**   | Confident, high-quality interface | Trust binding mode         |
| **0.6-0.8** | Uncertain region                  | Verify with experiments    |
| **\< 0.6**  | Low confidence                    | Likely incorrect interface |

<Warning>
  ipTM only considers inter-chain interactions. For single-chain predictions, use pTM instead.
</Warning>

### Types of ipTM Metrics

<Tabs>
  <Tab title="Global ipTM">
    Overall confidence in all interfaces:

    ```json theme={null}
    {
      "iptm": 0.72  // All chain-chain interfaces
    }
    ```

    Use for ranking predictions of the full complex.
  </Tab>

  <Tab title="Per-Chain ipTM">
    Average interface confidence per chain:

    ```json theme={null}
    {
      "chain_iptm": [0.85, 0.68, 0.91]
    }
    ```

    Element `i` = average ipTM between chain i and all other chains.

    **Use case**: Ranking predictions when interested in specific chain (e.g., ligand binding)
  </Tab>

  <Tab title="Pair-wise ipTM">
    Interface confidence for each chain pair:

    ```json theme={null}
    {
      "chain_pair_iptm": [
        [0.89, 0.72, 0.15],  // Chain A to A,B,C
        [0.72, 0.84, 0.20],  // Chain B to A,B,C
        [0.15, 0.20, 0.78]   // Chain C to A,B,C
      ]
    }
    ```

    * **Diagonal (i,i)**: pTM for chain i
    * **Off-diagonal (i,j)**: ipTM for i-j interface

    **Use case**: Antibody-antigen, known interacting pairs
  </Tab>
</Tabs>

### Ranking Strategies

<Steps>
  <Step title="Full Complex">
    Use global `iptm` or composite `ranking_score`
  </Step>

  <Step title="Specific Interface">
    Use `chain_pair_iptm[i][j]` for known interacting chains i and j
  </Step>

  <Step title="Specific Chain">
    Use `chain_iptm[i]` when you care about chain i's interactions but don't know partners
  </Step>

  <Step title="Chain Structure">
    Use `chain_ptm[i]` when only the structure of chain i matters
  </Step>
</Steps>

## Ranking Score

A composite score combining multiple metrics:

```text theme={null}
ranking_score = 0.8 × ipTM + 0.2 × pTM + 0.5 × disorder - 100 × has_clash
```

<Warning>
  **Use ranking\_score ONLY for ranking predictions**, not for absolute quality assessment.
</Warning>

### Components

<ParamField path="ipTM" type="number">
  Interface confidence (0-1) - weighted 0.8
</ParamField>

<ParamField path="pTM" type="number">
  Overall structure confidence (0-1) - weighted 0.2
</ParamField>

<ParamField path="disorder" type="number">
  Fraction disordered (0-1) - penalizes spurious helices (+0.5)
</ParamField>

<ParamField path="has_clash" type="boolean">
  Significant clashing atoms - strong penalty (-100)
</ParamField>

### When to Use

<Tabs>
  <Tab title="Use Ranking Score For">
    * Selecting best prediction from multiple samples
    * Comparing predictions with same input
    * Automated prediction pipelines
    * Initial filtering before manual inspection
  </Tab>

  <Tab title="Don't Use Ranking Score For">
    * Absolute quality assessment
    * Comparing different proteins
    * Scientific reporting (use pTM/ipTM)
    * Deciding if prediction is correct
  </Tab>
</Tabs>

## Additional Metrics

### Contact Probabilities

```json theme={null}
{
  "contact_probs": [[0.95, 0.12, ...],  // num_tokens × num_tokens
                     [0.12, 0.98, ...]]
}
```

Probability that two tokens are within 8Å (representative atoms).

**Use cases**:

* Identifying likely contacts before structure determination
* Assessing binding likelihood
* Filtering potential interfaces

### Chain-Pair PAE Min

```json theme={null}
{
  "chain_pair_pae_min": [
    [0.5, 15.2, 18.7],
    [15.1, 0.4, 19.2],
    [18.9, 19.5, 0.6]
  ]
}
```

Lowest PAE value from chain i to chain j.

<Note>
  **Correlates with binding**: Low `chain_pair_pae_min` between chains suggests they interact. Can distinguish binders from non-binders.
</Note>

### Disorder Fraction

```json theme={null}
{
  "fraction_disordered": 0.23
}
```

Fraction of structure classified as disordered (0-1), measured by accessible surface area.

**Interpretation**:

* High disorder: Flexible regions, loops, or prediction artifacts
* Used in ranking score to penalize spurious helices

### Clash Detection

```json theme={null}
{
  "has_clash": false
}
```

Boolean indicating significant clashing atoms:

* More than 50% of any chain has clashes, OR
* More than 100 clashing atoms in any chain

**Action**: If true, consider re-running with different seeds or inspect manually.

## Practical Guidelines

<Steps>
  <Step title="Initial Assessment">
    Check `ranking_score`, `ptm`, and `iptm` in summary JSON

    * pTM > 0.5: Likely correct fold
    * ipTM > 0.8: Confident interface
    * has\_clash = false: No major issues
  </Step>

  <Step title="Visual Inspection">
    Load structure colored by pLDDT (B-factor)

    * Blue regions (>90): High confidence
    * Red regions (\<50): Disordered/uncertain
  </Step>

  <Step title="Interface Analysis">
    Examine PAE plot

    * Low PAE blocks between chains: Good interface
    * High PAE: Uncertain interaction
  </Step>

  <Step title="Chain-Specific Quality">
    Check per-chain metrics

    * `chain_ptm[i]`: Structure of chain i
    * `chain_iptm[i]`: Chain i's interfaces
    * `chain_pair_pae_min[i][j]`: Likely interaction between i and j
  </Step>

  <Step title="Domain Analysis">
    For multi-domain proteins:

    * Low within-domain PAE: Well-structured domains
    * High between-domain PAE: Flexible linkers
  </Step>
</Steps>

## Additional Resources

<CardGroup cols={2}>
  <Card title="AlphaFold 3 Paper" icon="file-pdf" href="https://www.nature.com/articles/s41586-024-07487-w">
    Detailed description of confidence metrics
  </Card>

  <Card title="EBI Training" icon="graduation-cap" href="https://www.ebi.ac.uk/training/online/courses/alphafold/">
    Practical guide to confidence scores (AlphaFold 2 focused, still applicable)
  </Card>
</CardGroup>

## Next Steps

<CardGroup cols={2}>
  <Card title="Output Format" icon="folder-open" href="/guides/output-format">
    Learn about output file structure
  </Card>

  <Card title="Performance" icon="gauge" href="/guides/performance-optimization">
    Optimize predictions for better results
  </Card>
</CardGroup>