Merge pull request #9 from Syntist/updated-to-proteorift

Added changes to run on proteorift, more robust intsallation

.gitignore

@@ -17,4 +17,8 @@ __pycache__
 
 run_search.build
 run_search.dist
-run_search.onefile-build
+run_search.onefile-build
+ProteoRift-main
+config.ini
+env.json
+maestro-*

README.md

@@ -22,7 +22,9 @@ Download installer.bin from release
 Require Dependencies  >= Ubuntu 20.04 
 ```
 sudo apt install build-essential fuse
+```
 
+Now Run the Installer
 ```
 chmod +x installer.bin
 ./installer.bin

electron-app/src/main/index.ts

@@ -11,13 +11,13 @@ if (fs.existsSync(path.resolve(process.cwd(), '../../env.json'))) {
   configPath = path.resolve(process.cwd(), '../../env.json')
 } else if (fs.existsSync(path.resolve(process.cwd(), './env.json'))) {
   configPath = path.resolve(process.cwd(), './env.json')
+} else if (fs.existsSync(path.resolve(process.cwd(), '../env.json'))) {
+  configPath = path.resolve(process.cwd(), '../env.json')
 } else {
   console.log('env.json can not be found, exiting')
   process.exit()
 }
 
-console.log(configPath)
-
 const rawData = fs.readFileSync(configPath, 'utf-8')
 const config = JSON.parse(rawData)
 

electron-app/src/renderer/src/App.tsx

@@ -6,17 +6,17 @@ import { Runner } from './components/Runner'
 
 const specollate = `${await window.electron.getEnvVariable('SPECOLLATE')} -c ${await window.electron.getEnvVariable('SPECOLLATE_CONFIG')}`
 
-console.log(specollate)
+const proteorift = `${await window.electron.getEnvVariable('PROTEORIFT')} -c ${await window.electron.getEnvVariable('SPECOLLATE_CONFIG')}`
 
 function App(): JSX.Element {
   const [show, setShow] = useState(false)
 
   const runCommand = (gpu: boolean) => {
     setShow(true)
     if (gpu) {
-      window.electron.runCmd(`${specollate} -u gpu`)
+      window.electron.runCmd(`${proteorift} -u gpu`)
     } else {
-      window.electron.runCmd(specollate)
+      window.electron.runCmd(proteorift)
     }
   }
 

electron-app/src/renderer/src/components/ConfigForm/index.tsx

@@ -13,7 +13,7 @@ import {
   Checkbox
 } from '@mui/material'
 import FolderSelector from '../FolderSelector'
-import { configBuilder } from '@renderer/utils/helper'
+import { configBuilder, configBuilderV2 } from '@renderer/utils/helper'
 import { useState } from 'react'
 
 const ConfigForm = ({ run }: any) => {
@@ -44,7 +44,7 @@ const ConfigForm = ({ run }: any) => {
 
   console.log(errors)
   const onSubmit = (data) => {
-    configBuilder(data)
+    configBuilderV2(data)
     run()
   }
 

electron-app/src/renderer/src/utils/helper.tsx

@@ -1,5 +1,6 @@
-const MODEL = import.meta.env.VITE_MODEL
-const CONFIG = import.meta.env.VITE_CONFIG
+const MODEL = await window.electron.getEnvVariable('MODEL')
+const MODEL_2 = await window.electron.getEnvVariable('MODEL_2')
+const CONFIG = await window.electron.getEnvVariable('SPECOLLATE_CONFIG')
 
 export const configBuilder = async (data: any) => {
   const config = `
@@ -102,3 +103,136 @@ batch_size : 1024
 `
   await window.electron.ipcRenderer.invoke('write-file', CONFIG, config)
 }
+
+
+export const configBuilderV2 = async (data: any) => {
+  const config = `
+[preprocess]
+
+# For raptor
+# in_tensor_dir : ./data/train-ready/pred-full/
+
+# For comet
+# in_tensor_dir : /scratch/mtari008/37154933/pred-full-deepnovo/
+
+# For expanse
+in_tensor_dir : /lclhome/mtari008/job_2436627/nist_massiv_80k_ch_graymass/
+
+############ INPUT PARAMETERS ############
+[input]
+
+# file paths
+mgf_dir : /lclhome/mtari008/data/spectra/labeled/fruitfly.PXD004120
+prep_dir: sample_data/preprocess_files
+
+; val_dir : /lclhome/mtari008/data/deepatles/train_ready/nist-masive-deepnovo-5k-ch1-3-len7-30-200-mod-mass
+
+# The array size to store a spectrum.
+spec_size : 50000
+
+# Max charge value to be used to read spectrum files.
+charge : 5
+
+# Whether to use modifications or not.
+use_mods : True
+
+# Max mods per peptide
+num_mods: 5
+
+# Number of species the training dataset contains.
+num_species : 9 
+
+master_port : 12347
+
+rank : 1
+
+############ DATABASE SEARCH PARAMETERS ############
+[search]
+
+mgf_dir: ${data.mgf_dir}
+prep_path: ${data.prep_dir}
+pep_dir: ${data.pep_dir}
+out_pin_dir : ${data.out_pin_dir}
+
+model_name : ${MODEL_2}
+specollate_model_path: ${MODEL}
+
+
+# Batch sizes for forward pass through the network
+spec_batch_size : ${data.spec_batch_size}
+pep_batch_size : ${data.pep_batch_size}
+
+# Batch size for database search
+search_spec_batch_size : ${data.search_spec_batch_size}
+
+precursor_tolerance : ${data.precursor_tolerance} # Precursor tolerance to use during database search (Da or ppm)
+precursor_tolerance_type : ${data.precursor_tolerance_type} # either ppm or Da
+
+keep_psms : ${data.keep_psms} # Number of top scoring psms to keep
+
+# Number of modified peptides to be generated to search against. 
+# Different than the one in input section
+num_mods : ${data.num_mods}
+
+charge: ${data.charge} # charge to be used during search
+
+############ FILTERING PARAMETERS ############
+[filter]
+length_filter: True
+len_tol_neg: -1
+len_tol_pos: 1
+missed_cleavages_filter: True
+modification_filter: True
+
+############### OUT OF CORE PARAMETERS ##############
+[ooc]
+chunk_size: 10000000
+
+############ MACHINE LEARNING PARAMETERS ############
+[ml]
+
+batch_size : 1024
+
+test_size : 0.2
+
+max_spec_len : 200
+min_pep_len: 7
+max_pep_len : 30
+# slightly larger than max_pep_len to account for modifications
+pep_seq_len : 36
+max_clvs : 2
+embedding_dim : 1024
+encoder_layers : 4
+num_heads : 16
+
+train_count : 0
+
+ce_weight_clv : 1
+ce_weight_mod : 1
+mse_weight : 3
+
+dropout : 0.3
+
+lr : 0.0001
+
+weight_decay : 0.0001
+
+epochs : 5
+
+margin : 0.2
+
+read_split_listing : False
+
+############ DEFAULT VALUES ############
+# DO NOT CHANGE
+[default]
+msp_file : /data/human_consensus_final_true_lib.msp
+mgf_files : /data/
+spec_size : 8000
+charge : 2
+use_mods : False
+batch_size : 1024
+
+`
+  await window.electron.ipcRenderer.invoke('write-file', CONFIG, config)
+}

electron-app/tsconfig.web.json

@@ -4,7 +4,8 @@
     "src/renderer/src/env.d.ts",
     "src/renderer/src/**/*",
     "src/renderer/src/**/*.tsx",
-    "src/preload/*.d.ts"
+    "src/preload/*.d.ts",
+    "types"
   ],
   "compilerOptions": {
     "composite": true,

electron-app/types/electron-api.d.ts

@@ -0,0 +1,11 @@
+// electron-api.d.ts
+import { ElectronAPI } from '@electron-toolkit/preload'
+
+declare module '@electron-toolkit/preload' {
+  interface ElectronAPI {
+    getEnvVariable: (key: string) => string | undefined
+    runCmd: any
+    onCmdOutput: any
+    terminateCmd: any
+  }
+}

installer.py

@@ -6,8 +6,10 @@
 import requests
 import glob
 from zipfile import ZipFile
+import argparse
 
 MODEL_URL = "https://github.com/pcdslab/ProteoRift/releases/download/V1.0.0/specollate_model_weights.pt"
+MODEL_2_URL = "https://github.com/pcdslab/ProteoRift/releases/download/V1.0.0/proteorift_model_weights.pt"
 
 
 url = f'https://api.github.com/repos/pcdslab/MAESTRO/releases/latest'
@@ -67,6 +69,7 @@ def main():
     # Copy SpeCollate to electron-app
     electron_app_dir = Path(path)
     spe_collate_dest = Path(path) / "SpeCollate"
+    proteo_dest = Path(path) / "ProteoRift-main"
     app_name = ""
 
     if spe_collate_dest.exists():
@@ -78,6 +81,16 @@ def main():
         run_command(f"cd *-MAESTRO* && cp -r SpeCollate {electron_app_dir}")
         run_command(f"rm -rf {file}")
         run_command(f"rm -rf *-MAESTRO*")
+
+    if proteo_dest.exists():
+        print("ProteoRift Exist")
+    else:
+        print("ProteoRift Doesn't Exist, Downloading")
+        file = download_file("https://github.com/pcdslab/ProteoRift/archive/refs/heads/main.zip", electron_app_dir)
+        extract_zip(file, electron_app_dir)
+
+        print(file)
+        run_command(f"rm -rf {file}")
 
     if check_for_electron_app():
         app_name = check_for_electron_app()[0]
@@ -87,6 +100,7 @@ def main():
 
     # Download the model file
     model_filepath = download_file(MODEL_URL, models_dir)
+    model_2_filepath = download_file(MODEL_2_URL, models_dir)
 
     # Install Python if not installed
     python_bin = python_dir / "bin/python3"
@@ -115,17 +129,32 @@ def main():
     with env_file.open("a") as f:
         f.write("{\n")
         f.write(f'"SPECOLLATE":"{python_bin} {spe_collate_dest}/run_search.py",\n')
+        f.write(f'"PROTEORIFT":"{python_bin} {proteo_dest}/run_search.py",\n')
         f.write(f'"MODEL":"{model_filepath}",\n')
+        f.write(f'"MODEL_2":"{model_2_filepath}",\n')
         f.write(f'"SPECOLLATE_CONFIG":"{Path(path)}/config.ini"\n')
         f.write("}")
 
     print(f"Environment variables written to {env_file}")
 
     # Install npm dependencies and start the Electron app
     os.chdir(electron_app_dir)
-
-    run_command(f"chmod +x {electron_app_dir}/{app_name}" )
-    run_command(f"{electron_app_dir}/{app_name}")
+
+    parser = argparse.ArgumentParser(description="Script with --dev flag")
+    parser.add_argument(
+        '--dev', 
+        action='store_true', 
+        help='Run the script in development mode'
+    )
+
+    args = parser.parse_args()
+
+    if args.dev:
+        os.chdir(f"{electron_app_dir}/electron-app")
+        run_command(f"npm run dev" )
+    else:
+        run_command(f"chmod +x {electron_app_dir}/{app_name}" )
+        run_command(f"{electron_app_dir}/{app_name}")
 
 if __name__ == "__main__":
     main()