Use IonNTxPred to identify potential ion channel-targeting proteins directly from amino acid sequences provided in FASTA format. Upon submission, the tool automatically evaluates your input across the following four target classes:

• 🧩 Sodium (Na⁺) channel toxins
• 🧩 Potassium (K⁺) channel toxins
• 🧩 Calcium (Ca²⁺) channel toxins
• 🧩 Other ion channel toxins

1️⃣ Input:

• Paste sequences or upload a FASTA file.

2️⃣ IonNTxPred combines powerful machine learning methods to deliver accurate predictions. Select a model

• A protein language model (PLM) (ESM2-t33) trained on millions of sequences
• A hybrid model that integrates ESM2-t33 + BLAST-based prediction for enhanced accuracy. By default, the prediction uses the hybrid model to ensure optimal performance.

3️⃣ Threshold Setting
You can adjust the prediction threshold (ranging from 0 to 1) to fine-tune the model’s sensitivity and specificity according to your needs:

• A lower threshold (e.g., 0.1) increases sensitivity (fewer false negatives), but may yield more false positives.
• A higher threshold (e.g., 0.9) increases specificity (fewer false positives), but may miss some true positives.
• Default threshold: 0.3

🔍 Recommended thresholds: Based on our analysis, we suggest using the following optimized thresholds for best performance on each ion channel type. Click here to see Recommended thresholds.

📥 Simply upload or paste your sequences in FASTA format and click on submit button to begin.
📘 For more information on the input format, model details, and how to interpret results, please visit our Help & Documentation.