From fdf3b235b88746681c46ea12bcded76ecf8e1f76 Mon Sep 17 00:00:00 2001 From: Milot Mirdita Date: Sun, 18 Aug 2024 02:36:55 +0900 Subject: [PATCH] Fix pandas trying to guess dtypes and breaking on invalid job names when reading input https://twitter.com/SDuninHorkawicz/status/1824704638173516040 --- colabfold/batch.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/colabfold/batch.py b/colabfold/batch.py index 49f925e8..2c571af9 100644 --- a/colabfold/batch.py +++ b/colabfold/batch.py @@ -584,7 +584,7 @@ def get_queries( if input_path.is_file(): if input_path.suffix == ".csv" or input_path.suffix == ".tsv": sep = "\t" if input_path.suffix == ".tsv" else "," - df = pandas.read_csv(input_path, sep=sep) + df = pandas.read_csv(input_path, sep=sep, dtype=str) assert "id" in df.columns and "sequence" in df.columns queries = [ (seq_id, sequence.upper().split(":"), None)