Added examples, and updated README

README.md

@@ -103,6 +103,7 @@ Anytime you run diffusion you should be sure to activate this conda environment
 ```
 conda activate SE3nv
 ```
+Total setup should take less than 30 minutes on a standard desktop computer.
 Note: Due to the variation in GPU types and drivers that users have access to, we are not able to make one environment that will run on all setups. As such, we are only providing a yml file with support for CUDA 11.1 and leaving it to each user to customize it to work on their setups. This customization will involve changing the cudatoolkit and (possibly) the PyTorch version specified in the yml file.
 
 ---

examples/design_timbarrel.sh

@@ -3,9 +3,9 @@
 # We specify the output path
 # We tell RFdiffusion that we want to do scaffoldguided design, and that we are not making a binder to a target (just a monomer)
 # We provide a path to a directory of TIM barrel scaffolds, generated with the helper script
-# We generate 200 designs, with a reduced noise scale during inference of 0.5
+# We generate 10 designs, with a reduced noise scale during inference of 0.5
 # We sample additional length to increase diversity of the outputs. Specifically, we mask the loops and insert 0-5 residues (randomly sampled per-loop) into each loop
 # We add 0-5 residues (randomly sampled) to the N and the C-terminus
 # This will allow the generation of diverse TIM barrels with slightly different length helices and strands
 
-../scripts/run_inference.py inference.output_prefix=example_outputs/design_tim_barrel scaffoldguided.scaffoldguided=True scaffoldguided.target_pdb=False scaffoldguided.scaffold_dir=tim_barrel_scaffold/ inference.num_designs=200 denoiser.noise_scale_ca=0.5 denoiser.noise_scale_frame=0.5 scaffoldguided.sampled_insertion=0-5 scaffoldguided.sampled_N=0-5 scaffoldguided.sampled_C=0-5
+../scripts/run_inference.py inference.output_prefix=example_outputs/design_tim_barrel scaffoldguided.scaffoldguided=True scaffoldguided.target_pdb=False scaffoldguided.scaffold_dir=tim_barrel_scaffold/ inference.num_designs=10 denoiser.noise_scale_ca=0.5 denoiser.noise_scale_frame=0.5 scaffoldguided.sampled_insertion=0-5 scaffoldguided.sampled_N=0-5 scaffoldguided.sampled_C=0-5

examples/design_unconditional_w_contact_potential.sh

@@ -0,0 +1,7 @@
+#!/bin/bash
+# Here, we're making some unconditional designs, and also using the 'monomer_contacts' auxillary potential function
+# We specify the path for the outputs
+# We tell RFdiffusion that designs should be 100-200 residues in length (randomly sampled each design)
+# We generate 10 such designs
+
+../scripts/run_inference.py inference.output_prefix=example_outputs/design_unconditional_w_contact_potential 'contigmap.contigs=[100-200]' inference.num_designs=10 'potentials.guiding_potentials=["type:monomer_contacts,weight:0.05"]'

examples/design_unconditional_w_monomer_ROG.sh

@@ -0,0 +1,10 @@
+#!/bin/bash
+# Here, we're making some unconditional designs
+# We are also applying a Radius Of Gyration potential which encourages the design to be more compact
+# We specify the path for the outputs
+# We tell RFdiffusion that designs should be 100-200 residues in length (randomly sampled each design)
+# We generate 10 such designs
+# We use the monomer_ROG potential, with guide scale 2 and quadratic decay
+# Note that this potential is probably not necessary in this kind of case, but is provided as an example
+
+../scripts/run_inference.py inference.output_prefix=example_outputs/design_monomer_ROG_unconditional 'contigmap.contigs=[100-200]' inference.num_designs=10 'potentials.guiding_potentials=["type:monomer_ROG,weight:1,min_dist:5"]' potentials.guide_scale=2 potentials.guide_decay="quadratic"

setup.py

@@ -1,10 +1,10 @@
 from setuptools import setup, find_packages
 
 setup(name='rfdiffusion',
-      version='1.0.0',
+      version='1.1.0',
       description='RFdiffusion is an open source method for protein structure generation.',
       author='Rosetta Commons',
       url='https://github.com/RosettaCommons/RFdiffusion',
       scripts=["scripts/run_inference.py"],
       packages=find_packages(),
-      install_requires=['torch', 'se3-transformer'])
+      install_requires=['torch', 'se3-transformer'])