check seq length

app.py

@@ -40,6 +40,7 @@ if HF_TOKEN is None:
 
 PLOT_TARGET_POINTS = int(os.environ.get("PLOT_TARGET_POINTS", "1500"))
 SEARCH_MAX_RESULTS = int(os.environ.get("SEARCH_MAX_RESULTS", "20"))
+MAX_SEQUENCE_SIZE = 1_048_576  # 1MB in bytes - maximum allowed sequence input size
 
 # -----------------------------
 # Load pipeline (reloadable)
@@ -553,6 +554,17 @@ def predict(
             raise gr.Error("chrom is required when use_coords=True")
         if start is None or end is None or int(end) <= int(start):
             raise gr.Error("start/end must be set and end > start when use_coords=True")
+        
+        # Check sequence size before fetching from API: max 1MB
+        # Each base pair is typically 1 byte, so check region length
+        region_length = int(end) - int(start)
+        if region_length > MAX_SEQUENCE_SIZE:
+            raise gr.Error(
+                f"Requested genomic region is too large ({region_length:,} base pairs). "
+                f"Maximum allowed size is {MAX_SEQUENCE_SIZE:,} base pairs (1MB). "
+                f"Please select a smaller region."
+            )
+        
         inputs = {
             "chrom": chrom,
             "start": int(start),
@@ -563,9 +575,8 @@ def predict(
         if not seq or not seq.strip():
             raise gr.Error("seq is required when use_coords=False")
         seq_stripped = seq.strip()
-        # Check sequence size: max 1MB (1,048,576 bytes)
+        # Check sequence size: max 1MB
         # Each character is typically 1 byte, so check length
-        MAX_SEQUENCE_SIZE = 1_048_576  # 1MB in bytes
         if len(seq_stripped) > MAX_SEQUENCE_SIZE:
             raise gr.Error(
                 f"Sequence input is too large ({len(seq_stripped):,} characters). "

requirements.txt

@@ -1,9 +1,10 @@
-gradio==6.1.0
-kaleido==1.2.0
-matplotlib==3.10.8
-numpy==2.3.5
-plotly==6.5.0
-pyBigWig==0.3.24
-requests==2.32.5
+gradio>=4.0.0
+kaleido
+matplotlib
+numpy
+plotly
+pyBigWig
+pyfaidx
+requests
 torch
-transformers==4.57.3
+transformers>=4.41.0