import pandas as pd
from proteobench.io.parsing.parse_settings import get_open_source_tools
# this file contains utility functions for rendering the result table in tab1_results and tab4_display_results_submitted
# === Open Source Tools ===
# Loaded from proteobench/io/parsing/io_parse_settings/tool_metadata.toml
OPEN_SOURCE_TOOLS = get_open_source_tools()
[docs]
def add_open_source_column(df: pd.DataFrame) -> pd.DataFrame:
"""
Add an 'open_source' column indicating whether the software is open source.
Parameters
----------
df : pd.DataFrame
DataFrame containing a 'software_name' column.
Returns
-------
pd.DataFrame
Copy of the DataFrame with an 'open_source' column inserted after 'software_name'.
Open source tools are marked with '✅', others with an empty string.
"""
if "software_name" not in df.columns:
return df
df = df.copy()
df["open_source"] = df["software_name"].apply(
lambda x: "✅" if str(x).lower() in OPEN_SOURCE_TOOLS else ""
)
cols = df.columns.tolist()
cols.remove("open_source")
idx = cols.index("software_name") + 1
cols.insert(idx, "open_source")
return df[cols]
# === Table Color Constants ===
COLOR_IDENTIFIER = "#F0F2F6"
COLOR_PARAMETER = "#FFFFFF"
COLOR_RESULT = "#F0F2F6"
COLOR_TECHNICAL = "#FFFFFF"
COLOR_ADDITIONAL = "#F0F2F6"
def _get_style_js(bg_color: str):
"""
Generates JavaScript for styling cells with a background color.
Parameters
----------
bg_color : str
Hex color string to use as the background.
Returns
-------
JsCode
A JavaScript code block that defines the style.
"""
from st_aggrid import JsCode
return JsCode(
f"""
function(params) {{
return {{
'backgroundColor': '{bg_color}',
'color': 'black',
'fontWeight': 'normal'
}}
}}
"""
)
[docs]
def render_aggrid(df: pd.DataFrame, grid_options, key):
"""
Renders a DataFrame using AgGrid with specified grid options and a unique key.
Parameters
----------
df : pd.DataFrame
The DataFrame to display in the grid.
grid_options : dict
Configuration options for AgGrid.
key : Any
Unique identifier for the grid instance (AgGrid does not work with UUID keys).
Returns
-------
None
This function renders the grid in the Streamlit interface and does not return a value.
"""
# Calculate dynamic height based on number of rows
# Row height ~50px + header ~40px + padding
row_height = 50
header_height = 40
padding = 10
num_rows = len(df)
calculated_height = (num_rows * row_height) + header_height + padding
# Set min and max bounds for usability
min_height = 150
max_height = 800
dynamic_height = max(min_height, min(calculated_height, max_height))
from st_aggrid import AgGrid
AgGrid(
df,
gridOptions=grid_options,
theme="alpine",
fit_columns_on_grid_load=False,
height=dynamic_height,
allow_unsafe_jscode=True,
key=f"aggrid::{str(key)}", # AgGrid does not work with UUID keys
)
[docs]
def prepare_display_dataframe(df: pd.DataFrame, highlight_id: str | None) -> pd.DataFrame:
"""
Prepares the DataFrame for display, including column filtering, ordering,
row highlighting, and numeric formatting.
Parameters
----------
df : pd.DataFrame
The filtered dataset for display.
highlight_id : str or None
The ProteoBench ID to highlight (adds a marker in the 'selected' column).
Returns
-------
pd.DataFrame
A formatted and sorted DataFrame ready for rendering.
"""
df = df.copy()
if len(df) == 0:
return df
df["selected"] = df["id"].apply(lambda x: "➡️" if x == highlight_id else "")
try:
identifier_cols = ["selected", "id"]
parameter_cols = [
"software_name",
"software_version",
"search_engine",
"search_engine_version",
"ident_fdr_psm",
"ident_fdr_protein",
"ident_fdr_peptide",
"enable_match_between_runs",
"precursor_mass_tolerance",
"fragment_mass_tolerance",
"enzyme",
"allowed_miscleavages",
"min_peptide_length",
"max_peptide_length",
"fixed_mods",
"variable_mods",
"max_mods",
"min_precursor_charge",
"max_precursor_charge",
"quantification_method",
"protein_inference",
"abundance_normalization_ions",
"submission_comments",
]
result_cols = ["median_abs_epsilon", "mean_abs_epsilon", "nr_prec", "results"]
technical_cols = [
"proteobench_version",
"intermediate_hash",
"hover_text",
"color",
"old_new",
"is_temporary",
"comments",
"scatter_size",
]
# Define display column order
cols = identifier_cols + parameter_cols + result_cols + technical_cols
cols = [col for col in cols if col in df.columns]
additional_cols = [col for col in df.columns if col not in cols]
# remove boring columns
cols = [
col
for col in cols
if col not in ["comments", "scatter_size", "old_new", "is_temporary", "color", "hover_text"]
]
df = df[cols + additional_cols]
# Clean up values
df["results"] = df["results"].apply(str)
numeric_cols = df.select_dtypes(include=["float64", "int64"]).columns
df[numeric_cols] = df[numeric_cols].round(3)
df.sort_values(by="id", inplace=True)
except KeyError as e:
print(f"KeyError during DataFrame preparation: {e}")
return df
