support new AF-Multimer models

README.md

@@ -8,14 +8,19 @@ released by DeepMind in July 2021. After DeepMind released AlphaFold-Multimer [v
 
 ## Updates and Features
 
-#### Version 1.2 (2022-02-19)
+#### Version 1.2.2 (2022-03-10)
 
-- Added support to AF-Multimer neural network models in both paired and unpaired MSAs modes
+- Minor bug fix
+- Support AF-Multimer v2 neural network models (AF v2.2.0)
+
+#### Version 1.2.0 (2022-02-19)
+
+- Added support to AF-Multimer neural network models (AF v2.1.0) in both paired and unpaired MSAs modes
 - Domain cropping pre-generated full monomer features (in unpaired MSAs mode only)
 - Checkpoint option for model inference
 - Refactored code
 
-#### Version 1.0 (2021-11-09)
+#### Version 1.0.0 (2021-11-09)
 
 - Predicting structural models of a protein complex
 - Paired MSAs not required for complex modeling

example/README.md

@@ -4,7 +4,7 @@
 The main script for the example is the shell script `run_af2comp.sh`.
 First, use a text editor to open this script, and modify it to provide the correct
 python environment of AlphaFold, if not already loaded. Then, provide the correct path to the parameter files of AlphaFold neural network
-models. In AF's releases, these models are named as `model_x_ptm.npz` or `model_x_multimer.npz`. Note that AlphaFold-Multimer's models (version 2.1.1) are not required if you only wish to run the original monomer DL models.
+models. In AF's releases, these models are named as `params_model_x_ptm.npz` (AF version 2.0.x), `params_model_x_multimer.npz` (AF version 2.1.x), `params_model_x_multimer_v2.npz` (AF version 2.2.x). Note that AlphaFold-Multimer's models are not required if you only wish to run the original monomer DL models.
 
 Run the script with the command,
 ```sh

example/run_af2comp.sh

@@ -16,10 +16,12 @@ out_dir=af2c_mod # model output directory, $out_dir/$target
 ### This preset defined the number of recycles, ensembles, MSA cluster sizes (for monomer_ptm models)
 preset=super   # up to 20 recycles, 1 ensemble.
 
-### Choose neural network model(s) from ['model_1/2/3/4/5_multimer' and 'model_1/2/3/4/5_ptm']
+### Choose neural network model(s) from ['model_1/2/3/4/5_multimer', 'model_1/2/3/4/5_multimer_v2', or 'model_1/2/3/4/5_ptm']
 # Using AF2 multimer model released in version 2.1.1
 #model=model_1_multimer,model_2_multime,rmodel_3_multimer,model_4_multimer,model_5_multimer
-model=model_1_multimer,model_2_multimer
+# Using AF2 multimer model released in version 2.2.0
+#model=model_1_multimer_v2,model_2_multime_v2,rmodel_3_multimer_v2,model_4_multimer_v2,model_5_multimer_v2
+model=model_1_multimer_v2,model_2_multimer_v2
 # Using AF2 monomer_ptm model released in version 2.0.1
 #model=model_1_ptm,model_2_ptm,model_3_ptm,model_4_ptm,model_5_ptm