Glossary

Base Sequence

Your Project Base Sequence is the protein sequence that you start the project with. This may be a wild-type sequence that is found in nature or a previously engineered sequence.

Batch ID

Note: a batch of data is often the same as a round of data.

A batch of data consists of assay values measured under comparable experimental conditions. Each batch typically contains a set of controls prepared in a way consistent with the sample.

When scientists repeat or start a new experiment, they add a newly purified control (such as their wild type), which defines a new batch of data. That new batch contains your new controls as well as all the data points in your plate(s).

Candidate Sequence

A Candidate Sequence is a protein sequence generated by Cradle's machine learning models that is going to be built and tested in your wet lab. The models generate tens of thousands of sequences, but you will only build and test a specific selection, for example, the 96 sequences that are predicted to perform best.

Generated Candidate

A Generated Candidate is a sequence designed/generated by Cradle's machine learning models. The models will generate tens of thousands of sequences.

Zero-Shot Round

If you don't have historical experimental data, you can still run a zero-shot round. While the model won't have any assay data to learn from, it will still learn the evolutionary context of your protein of interest. Therefore, a zero-shot will provide a robust set of candidates whose lab testing will provide highly learnable data for Cradle's ML.

If you don't have historical experimental data, you can still run a zero-shot round. While the models won't have any assay data to learn from, they will still learn the evolutionary context of your protein of interest. Therefore, a zero-shot will provide a robust set of candidates whose lab testing will provide highly learnable data for Cradle's machine learning.