@@ -229,7 +229,7 @@ Now that we know Docker is functioning properly, we can begin building our Docke
2292291. Clone this repository and `cd` into it.
230230
231231 ```bash
232- git clone https://github.com/amorehead /DeepInteract
232+ git clone https://github.com/BioinfoMachineLearning /DeepInteract
233233 cd DeepInteract/
234234 DI_DIR=$(pwd)
235235 ```
@@ -281,7 +281,7 @@ First, install and configure Conda environment:
281281
282282```bash
283283# Clone this repository:
284- git clone https://github.com/amorehead /DeepInteract
284+ git clone https://github.com/BioinfoMachineLearning /DeepInteract
285285
286286# Change to project directory:
287287cd DeepInteract
@@ -372,7 +372,8 @@ To train, retrain, or cross-validate DeepInteract models using DIPS-Plus and/or
372372
373373``` bash
374374# Download and extract preprocessed DGLGraphs for DIPS-Plus and CASP-CAPRI
375- # Requires 7GB of free space
375+ # Requires ~55GB of free space
376+ mkdir -p project/datasets/DIPS/final
376377cd project/datasets/DIPS/final
377378
378379# Download DIPS-Plus
@@ -392,6 +393,7 @@ tar -xzf final_processed_dips.tar.gz
392393rm final_processed_dips.tar.gz.parta* final_raw_dips.tar.gz final_processed_dips.tar.gz
393394
394395# Download CASP-CAPRI
396+ mkdir -p ../../CASP_CAPRI/final
395397cd ../../CASP_CAPRI/final
396398wget https://zenodo.org/record/5546775/files/final_raw_casp_capri.tar.gz
397399wget https://zenodo.org/record/5546775/files/final_processed_casp_capri.tar.gz
@@ -402,6 +404,16 @@ tar -xzf final_processed_casp_capri.tar.gz
402404rm final_raw_casp_capri.tar.gz final_processed_casp_capri.tar.gz
403405```
404406
407+ Navigate to the project directory and run the training script with the parameters desired:
408+
409+ ``` bash
410+ cd project
411+ python3 lit_model_train.py --lr 1e-3 --weight_decay 1e-2
412+ cd ..
413+ ```
414+
415+ ## Inference
416+
405417### Download trained model checkpoint
406418
407419``` bash
@@ -413,15 +425,6 @@ mkdir -p project/checkpoints
413425wget -P project/checkpoints https://zenodo.org/record/5546775/files/LitGINI-GeoTran-DilResNet.ckpt
414426```
415427
416- Navigate to the project directory and run the training script with the parameters desired:
417-
418- ``` bash
419- cd project
420- python3 lit_model_train.py --lr 1e-3 --weight_decay --1e-2
421- cd ..
422- ```
423-
424- ## Inference
425428### Predict interface contact probability maps
426429
427430Navigate to the project directory and run the prediction script
0 commit comments