Update readme and the msa selection service.

THU-ATOM · Jun 25, 2024 · 9053d88 · 9053d88
1 parent b9a2c4c
commit 9053d88
Showing 6 changed files with 38 additions and 25 deletions.
diff --git a/README.md b/README.md
@@ -9,7 +9,9 @@
 - Flower for monitoring the Celery tasks
 
 ## Introduction
-AIRFold is 
+![AIRFold Framework](imgs/Frame.png)
+
+AIRFold is an open-source platform for protein structure prediction.
 
 ## Quick Start
 
@@ -103,25 +105,32 @@ Please follow these steps:
 ## Command for different functions
 
 ### Multiple sequence alignment generation
-
-    ```bash
-    python run_mode.py --input_path example.fasta --mode msa
-    ```
+```bash
+# Input: Protein sequences in fasta format.
+# Output: Multiple sequence alignment results in a3m format.
+python run_mode.py --input_path example.fasta --mode msa
+```
 ### Pretrained embedding generation
 
-    ```bash
-    python run_mode.py --input_path example.fasta --mode feature
-    ```
+```bash
+# Input: Protein sequences in fasta format.
+# Output: Generated sequence embeddings in pickle format.
+python run_mode.py --input_path example.fasta --mode feature
+```
 ### Protein contact map prediction
 
-    ```bash
-    python run_mode.py --input_path example.fasta --mode disgram
-    ```
+```bash
+# Input: Protein sequences in fasta format.
+# Output: Generated contact map in pickle format.
+python run_mode.py --input_path example.fasta --mode disgram
+```
 
 ### Protein structure prediction
 
-    ```bash
-    python run_mode.py --input_path example.fasta --mode pipline
-    ```
+```bash
+# Input: Protein sequences in fasta format.
+# Output: Protein structure in pdb format.
+python run_mode.py --input_path example.fasta --mode pipline
+```
 
 
diff --git a/batch_run.py b/batch_run.py
@@ -124,7 +124,7 @@ def main():
     # json_file = argv.input_path
     # with open("./tmp/temp_5000_128_1_mmseqs.json", 'r') as jf:
     #     request_dict = json.load(jf)
-    with open("./tmp/temp_10000_128_1_plmsim.json", 'r') as jf:
+    with open("./tmp/temp_6000_64_1_plmsim.json", 'r') as jf:
         request_dict = json.load(jf)
 
     # weeks = ['2024.02.17', '2024.02.24', '2024.03.02', '2024.03.09', 

diff --git a/gui/stats.html b/gui/stats.html
diff --git a/imgs/Frame.png b/imgs/Frame.png
diff --git a/services/alphafold/worker.py b/services/alphafold/worker.py
@@ -285,7 +285,7 @@ def run(self):
         key_list = list(str_dict.keys())
         msa_paths = []
         for idx in range(len(key_list)):
-            selected_msa_path = ptree.strategy.strategy_list[idx] + "_dp.a3m"
+            selected_msa_path = str(ptree.strategy.strategy_list[idx]) + "_dp.a3m"
             msa_paths.append(str(selected_msa_path))
 
         # get selected_template_feat
@@ -428,9 +428,10 @@ def run(self):
         # get msa_path
         str_dict = misc.safe_get(self.requests[0], ["run_config", "msa_select"])
         key_list = list(str_dict.keys())
-        for index in range(len(key_list)):
-            selected_msa_path = ptree.strategy.strategy_list[index]
-        msa_paths = [str(selected_msa_path)]
+        msa_paths = []
+        for idx in range(len(key_list)):
+            selected_msa_path = str(ptree.strategy.strategy_list[idx]) + "_dp.a3m"
+            msa_paths.append(str(selected_msa_path))
 
         # get selected_template_feat
         selected_template_feat_path = str(ptree.alphafold.selected_template_feat)

diff --git a/services/selectmsa/worker.py b/services/selectmsa/worker.py
@@ -142,15 +142,18 @@ def build_command(self, request: Dict[str, Any]) -> list:
         for idx in range(len(key_list)):
             """TODO current implementation could not be compatible with ABA-like strategy"""
 
-            method_ = select_args[key_list[idx]]
+            method_ = key_list[idx]
+            logger.info(f"The Method for msa selection: {method_}")
             executed_file = (Path(__file__).resolve().parent / "lib" / "strategy" / f"{method_}.py")
             input_fasta_path = str(ptree.seq.fasta)
-            self.output_prefix = ptree.strategy.strategy_list[idx]
-            self.output_prefix.parent.mkdir(exist_ok=True, parents=True)
+            output_prefix = ptree.strategy.strategy_list[idx]
+            output_prefix.parent.mkdir(exist_ok=True, parents=True)
+            self.output_prefix = str(output_prefix)
 
-            for tag in select_args[method_]["least_seqs"]:
+            ls_dict = misc.safe_get(select_args, [method_, "least_seqs"])
+            for tag in ls_dict.keys():
 
-                least_seqs = select_args[method_]["least_seqs"][tag]
+                least_seqs = ls_dict[tag]
                 if tag == "hj" and "hhblits" in search_args.keys() and "jackhmmer" in search_args.keys():
                     input_path = str(ptree.search.integrated_search_hj_a3m_dp)
                 if tag == "bl" and "blast" in search_args.keys():