update readme

README.md

@@ -518,48 +518,48 @@ print(predictor.predict_property("nf",    "smiles",     smiles, uncertainty=True
 You can also find the same description in the paper or in the PeptiVerse app `Documentation` tab.
 
 ---
-#### 🩸 Hemolysis Prediction
-50% of read blood cells being lysed at x ug/ml concetration (HC50). If HC50 < 100uM, considered as hemolytic, otherwise non-hemolytic, resulting in a binary 0/1 dataset. The predicted probability should therefore be interpreted as a risk indicator, not an exact concentration estimate.
-**Output interpretation:**
+### 🩸 Hemolysis Prediction<br>
+50% of read blood cells being lysed at x ug/ml concetration (HC50). If HC50 < 100uM, considered as hemolytic, otherwise non-hemolytic, resulting in a binary 0/1 dataset. The predicted probability should therefore be interpreted as a risk indicator, not an exact concentration estimate.<br>
+**Output interpretation:**<br>
 - Score close to 1.0 = high probability of red blood cell membrane disruption
 - Score close to 0.0 = non-hemolytic
 ---
 
-#### 💧 Solubility Prediction
-Outputs a probability (0–1) that a peptide remains soluble in aqueous conditions.
-**Output interpretation:**
+### 💧 Solubility Prediction<br>
+Outputs a probability (0–1) that a peptide remains soluble in aqueous conditions.<br>
+**Output interpretation:**<br>
 - Score close to 1.0 = highly soluble
 - Score close to 0.0 = poorly soluble
 
 ---
 
-#### 👯 Non-Fouling Prediction
-Higher scores indicate stronger non-fouling behavior, desirable for circulation and surface-exposed applications.
-**Output interpretation:**
+### 👯 Non-Fouling Prediction<br>
+Higher scores indicate stronger non-fouling behavior, desirable for circulation and surface-exposed applications.<br>
+**Output interpretation:**<br>
 - Score close to 1.0 = non-fouling
 - Score close to 0.0 = fouling
 ---
 
-#### 🪣 Permeability Prediction
-Predicts membrane permeability on a log P scale.
-**Output interpretation:**
+### 🪣 Permeability Prediction<br>
+Predicts membrane permeability on a log P scale.<br>
+**Output interpretation:**<br>
 - Higher values = more permeable (>-6.0)
 - For penetrance predictions, it is a classification prediction, so within the [0, 1] range, closer to 1 indicates more permeable.
 ---
 
-#### ⏱️ Half-Life Prediction
-**Interpretation:** Predicted values reflect relative peptide stability for the unit in hours. Higher scores indicate longer persistence in serum, while lower scores suggest faster degradation.
+### ⏱️ Half-Life Prediction<br>
+**Interpretation:** Predicted values reflect relative peptide stability for the unit in hours. Higher scores indicate longer persistence in serum, while lower scores suggest faster degradation.<br>
 
 ---
 
-#### ☠️ Toxicity Prediction
-**Interpretation:** Outputs a probability (0–1) that a peptide exhibits toxic effects. Higher scores indicate increased toxicity risk.
+### ☠️ Toxicity Prediction<br>
+**Interpretation:** Outputs a probability (0–1) that a peptide exhibits toxic effects. Higher scores indicate increased toxicity risk.<br>
 
 ---
 
-#### 🔗 Binding Affinity Prediction
+### 🔗 Binding Affinity Prediction <br>
 
-Predicts peptide-protein binding affinity. Requires both peptide and target protein sequence.
+Predicts peptide-protein binding affinity. Requires both peptide and target protein sequence.<br>
 
 **Interpretation:**<br>
     - Scores ≥ 9 correspond to tight binders (K ≤ 10⁻⁹ M, nanomolar to picomolar range)<br>
@@ -569,15 +569,15 @@ Predicts peptide-protein binding affinity. Requires both peptide and target prot
     
 ---
 
-### Uncertainty Interpretation 
-#### Entropy (classifiers)
+### Uncertainty Interpretation <br>
+#### Entropy (classifiers)<br>
  
-Binary predictive entropy of the output probability $\bar{p}$:
+Binary predictive entropy of the output probability $\bar{p}$:<br>
  
-$$\mathcal{H} = -\bar{p}\log\bar{p} - (1 - \bar{p})\log(1 - \bar{p})$$
+$$\mathcal{H} = -\bar{p}\log\bar{p} - (1 - \bar{p})\log(1 - \bar{p})$$<br>
  
-- For **DNN classifiers**: $\bar{p}$ is the mean probability across 5 independently seeded models (deep ensemble). High entropy reflects both epistemic uncertainty (seed disagreement) and aleatoric uncertainty (collectively diffuse predictions).
-- For **XGBoost / SVM / ElasticNet classifiers**: $\bar{p}$ is the single model's output probability (or sigmoid of decision function for ElasticNet). Entropy reflects output confidence of a single model only.
+- For **DNN classifiers**: $\bar{p}$ is the mean probability across 5 independently seeded models (deep ensemble). High entropy reflects both epistemic uncertainty (seed disagreement) and aleatoric uncertainty (collectively diffuse predictions).<br>
+- For **XGBoost / SVM / ElasticNet classifiers**: $\bar{p}$ is the single model's output probability (or sigmoid of decision function for ElasticNet). Entropy reflects output confidence of a single model only.<br>
  
 | Range | Interpretation |
 |---|---|
@@ -589,33 +589,33 @@ $$\mathcal{H} = -\bar{p}\log\bar{p} - (1 - \bar{p})\log(1 - \bar{p})$$
  
 ---
  
-#### Adaptive Conformal Prediction Interval (regressors)
+#### Adaptive Conformal Prediction Interval (regressors)<br>
  
-Returned as a tuple `(lo, hi)` with 90% marginal coverage guarantee.
+Returned as a tuple `(lo, hi)` with 90% marginal coverage guarantee.<br>
  
-We implement the **residual normalised conformity score** following [Lei et al. (2018)](https://doi.org/10.1080/01621459.2017.1307116) and [Cordier et al. (2023) / MAPIE](https://proceedings.mlr.press/v204/cordier23a.html). An auxiliary XGBoost model $\hat{\sigma}(\mathbf{x})$ is trained on held-out embeddings and absolute residuals $|y_i - \hat{y}_i|$. At inference:
+We implement the **residual normalised conformity score** following [Lei et al. (2018)](https://doi.org/10.1080/01621459.2017.1307116) and [Cordier et al. (2023) / MAPIE](https://proceedings.mlr.press/v204/cordier23a.html). An auxiliary XGBoost model $\hat{\sigma}(\mathbf{x})$ is trained on held-out embeddings and absolute residuals $|y_i - \hat{y}_i|$. At inference:<br>
  
 $$[\hat{y}(\mathbf{x}) - q \cdot \hat{\sigma}(\mathbf{x}),\ \hat{y}(\mathbf{x}) + q \cdot \hat{\sigma}(\mathbf{x})]$$
  
-where $q$ is the $\lceil(n+1)(1-\alpha)\rceil / n$ quantile of the normalized scores $s_i = |y_i - \hat{y}_i| / \hat{\sigma}(\mathbf{x}_i)$.
+where $q$ is the $\lceil(n+1)(1-\alpha)\rceil / n$ quantile of the normalized scores $s_i = |y_i - \hat{y}_i| / \hat{\sigma}(\mathbf{x}_i)$.<br>
  
-- **Interval width varies per input** -- molecules more dissimilar to training data tend to receive wider intervals
-- **Coverage guarantee**: on exchangeable data, $P(y \in [\hat{y} - q\hat{\sigma},\ \hat{y} + q\hat{\sigma}]) \geq 0.90$
-- **The guarantee is marginal**, not conditional, as an unusually narrow interval on an out-of-distribution molecule does not guarantee correctness
-- **Full access**: we already computed MAPIE for all regression models, users are allowed to directly use them for customized model lists.
+- **Interval width varies per input** -- molecules more dissimilar to training data tend to receive wider intervals<br>
+- **Coverage guarantee**: on exchangeable data, $P(y \in [\hat{y} - q\hat{\sigma},\ \hat{y} + q\hat{\sigma}]) \geq 0.90$<br>
+- **The guarantee is marginal**, not conditional, as an unusually narrow interval on an out-of-distribution molecule does not guarantee correctness<br>
+- **Full access**: We already computed MAPIE for all regression models; users are allowed to directly use them for customized model lists.<br>
  
 ---
  
-#### Generating a MAPIE Bundle for a New Model
+#### Generating a MAPIE Bundle for a New Model<br>
  
-To enable conformal uncertainty for a newly trained regression model:
+To enable conformal uncertainty for a newly trained regression model:<br>
  
 ```bash
 # Fit adaptive conformal bundle from val_predictions.csv
 python fit_mapie_adaptive.py --root training_classifiers --prop <property_name>
 ```
  
-The script reads `sequence`/`smiles` and `y_pred`/`y_true` columns from the CSV, recomputes embeddings, fits the XGBoost $\hat{\sigma}$ model, and saves `mapie_calibration.joblib` into the model directory. The bundle is automatically detected and loaded by `PeptiVersePredictor` on next initialisation.
+The script reads `sequence`/`smiles` and `y_pred`/`y_true` columns from the CSV, recomputes embeddings, fits the XGBoost $\hat{\sigma}$ model, and saves `mapie_calibration.joblib` into the model directory. The bundle is automatically detected and loaded by `PeptiVersePredictor` on the next initialization.<br>