The challenge dataset stems from a high-density peptide microarray experiment that aimed to address how cross-reactive 8 different commercially available snake antivenoms are and where in the toxin sequence the antibodies they contain bind the toxin (epitope).

Each row in the dataset represents a k-mer (16 amino acid sequence within the toxin) and it has a signal column coming from the high-density peptide microarray experiment.

You will need to predict the signal column generated by a given Toxin_K_mer and Antivenom. You can use any other column available in the test set to enhance your predictions or enrich your data. We also facilitate the protein prot_bert embeddings for each row.

Watch this video, it is a walk-through of the starter notebook and the relevance of the challenge.

You can also view it here.

The data you are presented with includes the following columns:

1. ID : Unique identifier for each row
2. Toxin_UniprotID= identifier for a specific toxin
3. Position_start = The start position in the toxin global sequence of the k-mer Position_end = The end position in the toxin global sequence a given k-mer
4. Antivenom = name of the antivenom tested in the high-density peptide microarray experiment
5. Toxin_K_mer = string of 16 amino acids (16-mer) from a given toxin sequence
6. Signal = (target) The output of the experiment. A proxy for antivenom activity.
7. Genus = genus of snake the toxin stems from, e.g. Naja (cobra)
8. Species = species of snake the toxin originates from e.g. Naja nigricollis (Black-necked spitting cobra)
9. ProteinFam= Toxin protein family, e.g. three finger toxin (3FTx)
10. ProteinSubFam= Toxin sub-family, e.g. cytotoxin (a type of 3FTx)
11. ProteinSubSubFam= Toxin sub-sub-family, e.g. cytotoxin IA (a type of cytotoxin)