PredictModel: Predict structures of chains in a sequence file



Predict model can be used to generate predicted models from a sequence file. One predicted model is generated for each unique sequence.

Sequence file

The sequence file that you supply specifies what is going to be predicted and how many copies of each chain are present in the structure (one for every copy in the sequence file). All models that are created and input will be associated with one or more of the chains in your sequence file based on sequence identity (normally every model should match a chain in the sequence file exactly).

Model prediction

Predict model normally uses either a Phenix server or a Google Colab notebook to carry out AlphaFold prediction of one chain at a time.

The general procedure is to create a set of files that specify all the necessary inputs for model prediction (a sequence file, optional multiple sequence alignment file, optional templates, keywords for model prediction such as the number of models to generate, random seed, whether to use a multiple sequence alignment) and to package all these files in a single zipped tar (.tgz) file.

If Colab is specified, the Phenix GUI will provide a button that opens Google Colab in a browser, then you start the Colab notebook and upload the zipped tar file. The notebook runs the prediction and downloads a .zip file with the results. The Phenix GUI recognizes this file, opens it and puts the new prediction in the working directory. This directory is specified by the unique jobname that you provide for your work. You will need to be logged in to a Google account for this to work.

If the Phenix server is specified, the Phenix GUI will use the Phenix server to carry out the prediction and put it in the working directory. You will need to have the address of the Phenix server for this to work.

When prediction is being carried out, the Phenix GUI waits for the predicted models to appear in the working directory, then it writes out the resulting models.


Standard run of predict_model

Running predict_model is easy. From the command-line you can type:

phenix.predict_and_build jobname=myjob seq_file=seq.dat \
  prediction_server=PhenixServer \
  stop_after_predict=True rest_server.url=http://my_rest_server:2320/

Server problems

The most common problem in running Predict model is that either the Phenix server or the Colab server is not working as expected. Normally the first thing to try is just let the program retry (it will do this for a while normally). If that is not working, you can stop the program (abort in the GUI), and run again with the same parameters (i.e., the same jobname) except changing the server from PhenixServer to Colab or vice-versa. That will give you another chance. If neither server is working, you can take the files in the packaged .tgz file (listed in the GUI output), use them to get your own prediction with any server, and put the resulting predicted models in the place specified in the GUI or program output.

Specific limitations and problems:


Additional information

List of all available keywords