masster 0.3.13__py3-none-any.whl → 0.3.15__py3-none-any.whl

masster/study/plot.py

@@ -37,7 +37,6 @@ def plot_alignment(self, maps: bool = True, filename: str | None = None, width:
     # Local imports so the module can be used even if bokeh isn't needed elsewhere
     from bokeh.models import ColumnDataSource, HoverTool
     from bokeh.plotting import figure, show, output_file
-    from bokeh.palettes import Turbo256
     import pandas as pd
 
     # Build the before/after tabular data used for plotting
@@ -136,34 +135,99 @@ def plot_alignment(self, maps: bool = True, filename: str | None = None, width:
             self.logger.error("Column 'rt_original' not found in features_df. Alignment may not have been performed.")
             return
 
-        features_pd = self.features_df.to_pandas()
+        # Use Polars instead of pandas
+        features_df = self.features_df
 
-        sample_col = 'sample_uid' if 'sample_uid' in features_pd.columns else 'sample_name'
-        if sample_col not in features_pd.columns:
+        sample_col = 'sample_uid' if 'sample_uid' in features_df.columns else 'sample_name'
+        if sample_col not in features_df.columns:
             self.logger.error("No sample identifier column found in features_df.")
             return
 
-        samples = features_pd[sample_col].unique()
+        # Get unique samples using Polars
+        samples = features_df.select(pl.col(sample_col)).unique().to_series().to_list()
 
         for sample_idx, sample in enumerate(samples):
-            sample_data = features_pd[features_pd[sample_col] == sample]
-            max_inty = sample_data['inty'].max() if sample_data['inty'].max() > 0 else 1
+            # Filter sample data using Polars
+            sample_data = features_df.filter(pl.col(sample_col) == sample)
+            
+            # Calculate max intensity using Polars
+            max_inty = sample_data.select(pl.col('inty').max()).item()
+            max_inty = max_inty if max_inty and max_inty > 0 else 1
+            
             sample_name = str(sample)
-            sample_uid = sample if sample_col == 'sample_uid' else (sample_data['sample_uid'].iloc[0] if 'sample_uid' in sample_data.columns else sample)
-
-            for _, row in sample_data.iterrows():
-                before_data.append({'rt': row['rt_original'], 'mz': row['mz'], 'inty': row['inty'], 'alpha': row['inty'] / max_inty, 'sample_idx': sample_idx, 'sample_name': sample_name, 'sample_uid': sample_uid, 'size': markersize + 2 if sample_idx == 0 else markersize})
-                after_data.append({'rt': row['rt'], 'mz': row['mz'], 'inty': row['inty'], 'alpha': row['inty'] / max_inty, 'sample_idx': sample_idx, 'sample_name': sample_name, 'sample_uid': sample_uid, 'size': markersize + 2 if sample_idx == 0 else markersize})
-
-    # Ensure dataframes exist even if empty
-    before_df = pd.DataFrame(before_data)
-    after_df = pd.DataFrame(after_data)
-
-    # Create ColumnDataSources (safe even for empty dfs)
-    from bokeh.models import ColumnDataSource
+            # Get sample_uid - if sample_col is 'sample_uid', use sample directly
+            if sample_col == 'sample_uid':
+                sample_uid = sample
+            else:
+                # Try to get sample_uid from the first row if it exists
+                if 'sample_uid' in sample_data.columns:
+                    sample_uid = sample_data.select(pl.col('sample_uid')).item()
+                else:
+                    sample_uid = sample
+
+            # Convert to dict for iteration - more efficient than row-by-row processing
+            sample_dict = sample_data.select(['rt_original', 'rt', 'mz', 'inty']).to_dicts()
+            
+            for row_dict in sample_dict:
+                rt_original = row_dict['rt_original']
+                rt_current = row_dict['rt']
+                mz = row_dict['mz']
+                inty = row_dict['inty']
+                alpha = inty / max_inty
+                size = markersize + 2 if sample_idx == 0 else markersize
+                
+                before_data.append({
+                    'rt': rt_original, 'mz': mz, 'inty': inty, 'alpha': alpha, 
+                    'sample_idx': sample_idx, 'sample_name': sample_name, 
+                    'sample_uid': sample_uid, 'size': size
+                })
+                after_data.append({
+                    'rt': rt_current, 'mz': mz, 'inty': inty, 'alpha': alpha, 
+                    'sample_idx': sample_idx, 'sample_name': sample_name, 
+                    'sample_uid': sample_uid, 'size': size
+                })
+
+    # Get sample colors from samples_df using sample indices
+    # Extract unique sample information from the dictionaries we created
+    if before_data:
+        # Create mapping from sample_idx to sample_uid more efficiently
+        sample_idx_to_uid = {}
+        for item in before_data:
+            if item['sample_idx'] not in sample_idx_to_uid:
+                sample_idx_to_uid[item['sample_idx']] = item['sample_uid']
+    else:
+        sample_idx_to_uid = {}
+    
+    # Get colors from samples_df
+    sample_uids_list = list(sample_idx_to_uid.values())
+    if sample_uids_list and hasattr(self, 'samples_df') and self.samples_df is not None:
+        sample_colors = (
+            self.samples_df
+            .filter(pl.col("sample_uid").is_in(sample_uids_list))
+            .select(["sample_uid", "sample_color"])
+            .to_dict(as_series=False)
+        )
+        uid_to_color = dict(zip(sample_colors["sample_uid"], sample_colors["sample_color"]))
+    else:
+        uid_to_color = {}
 
-    before_source = ColumnDataSource(before_df)
-    after_source = ColumnDataSource(after_df)
+    # Create color map for sample indices
+    color_map: dict[int, str] = {}
+    for sample_idx, sample_uid in sample_idx_to_uid.items():
+        color_map[sample_idx] = uid_to_color.get(sample_uid, "#1f77b4")  # fallback to blue
+
+    # Add sample_color to data dictionaries before creating DataFrames
+    if before_data:
+        for item in before_data:
+            item['sample_color'] = color_map.get(item['sample_idx'], '#1f77b4')
+    
+    if after_data:
+        for item in after_data:
+            item['sample_color'] = color_map.get(item['sample_idx'], '#1f77b4')
+    
+    # Now create DataFrames with the sample_color already included
+    before_df = pd.DataFrame(before_data) if before_data else pd.DataFrame()
+    after_df = pd.DataFrame(after_data) if after_data else pd.DataFrame()
 
     # Create Bokeh figures
     p1 = figure(width=width, height=height, title='Original RT', x_axis_label='Retention Time (s)', y_axis_label='m/z', tools='pan,wheel_zoom,box_zoom,reset,save')
@@ -177,15 +241,9 @@ def plot_alignment(self, maps: bool = True, filename: str | None = None, width:
     p2.background_fill_color = 'white'
     p2.border_fill_color = 'white'
     p2.min_border = 0
-
-    # Color mapping using Turbo256
-    unique_samples = sorted(list(set(before_df['sample_idx'].tolist()))) if not before_df.empty else []
-    colors = Turbo256
-    color_map: dict[int, str] = {}
-    n = max(1, len(unique_samples))
-    step = max(1, 256 // n)
-    for i, sample_idx in enumerate(unique_samples):
-        color_map[sample_idx] = colors[(i * step) % 256]
+    
+    # Get unique sample indices for iteration
+    unique_samples = sorted(list(set(item['sample_idx'] for item in before_data))) if before_data else []
 
     renderers_before = []
     renderers_after = []
@@ -206,10 +264,10 @@ def plot_alignment(self, maps: bool = True, filename: str | None = None, width:
             renderers_after.append(r)
 
     # Add hover tools
-    hover1 = HoverTool(tooltips=[('Sample UID', '@sample_uid'), ('Sample Name', '@sample_name'), ('RT', '@rt{0.00}'), ('m/z', '@mz{0.0000}'), ('Intensity', '@inty{0.0e+0}')], renderers=renderers_before)
+    hover1 = HoverTool(tooltips=[('Sample UID', '@sample_uid'), ('Sample Name', '@sample_name'), ('Sample Color', '$color[swatch]:sample_color'), ('RT', '@rt{0.00}'), ('m/z', '@mz{0.0000}'), ('Intensity', '@inty{0.0e0}')], renderers=renderers_before)
     p1.add_tools(hover1)
 
-    hover2 = HoverTool(tooltips=[('Sample UID', '@sample_uid'), ('Sample Name', '@sample_name'), ('RT', '@rt{0.00}'), ('m/z', '@mz{0.0000}'), ('Intensity', '@inty{0.0e+0}')], renderers=renderers_after)
+    hover2 = HoverTool(tooltips=[('Sample UID', '@sample_uid'), ('Sample Name', '@sample_name'), ('Sample Color', '$color[swatch]:sample_color'), ('RT', '@rt{0.00}'), ('m/z', '@mz{0.0000}'), ('Intensity', '@inty{0.0e0}')], renderers=renderers_after)
     p2.add_tools(hover2)
 
     # Create layout with both plots side by side
@@ -232,13 +290,13 @@ def plot_consensus_2d(
     self,
     filename=None,
     colorby="number_samples",
+    cmap=None,
+    markersize=4,
     sizeby="inty_mean",
-    markersize=6,
-    size="dynamic",
+    scaling="dynamic",
     alpha=0.7,
-    cmap=None,
-    width=900,
-    height=900,
+    width=600,
+    height=450,
     mz_range=None,
     rt_range=None,
 ):
@@ -317,7 +375,7 @@ def plot_consensus_2d(
         ])
 
     if cmap is None:
-        cmap = "vi"
+        cmap = "viridis"
     elif cmap == "grey":
         cmap = "Greys256"
 
@@ -334,12 +392,49 @@ def plot_consensus_2d(
     except ImportError:
         from bokeh.models.annotations import ColorBar
     from bokeh.palettes import viridis
+    
+    # Import cmap for colormap handling
+    from cmap import Colormap
 
     # Convert Polars DataFrame to pandas for Bokeh compatibility
     data_pd = data.to_pandas()
     source = ColumnDataSource(data_pd)
+    
+    # Handle colormap using cmap.Colormap
+    try:
+        # Get colormap palette using cmap
+        if isinstance(cmap, str):
+            colormap = Colormap(cmap)
+            # Generate 256 colors and convert to hex
+            import numpy as np
+            import matplotlib.colors as mcolors
+            colors = colormap(np.linspace(0, 1, 256))
+            palette = [mcolors.rgb2hex(color) for color in colors]
+        else:
+            colormap = cmap
+            # Try to use to_bokeh() method first
+            try:
+                palette = colormap.to_bokeh()
+                # Ensure we got a color palette, not another mapper
+                if not isinstance(palette, (list, tuple)):
+                    # Fall back to generating colors manually
+                    import numpy as np
+                    import matplotlib.colors as mcolors
+                    colors = colormap(np.linspace(0, 1, 256))
+                    palette = [mcolors.rgb2hex(color) for color in colors]
+            except AttributeError:
+                # Fall back to generating colors manually
+                import numpy as np
+                import matplotlib.colors as mcolors
+                colors = colormap(np.linspace(0, 1, 256))
+                palette = [mcolors.rgb2hex(color) for color in colors]
+    except (AttributeError, ValueError, TypeError) as e:
+        # Fallback to viridis if cmap interpretation fails
+        self.logger.warning(f"Could not interpret colormap '{cmap}': {e}, falling back to viridis")
+        palette = viridis(256)
+    
     color_mapper = LinearColorMapper(
-        palette=viridis(256),
+        palette=palette,
         low=data[colorby].min(),
         high=data[colorby].max(),
     )
@@ -352,11 +447,11 @@ def plot_consensus_2d(
     p.xaxis.axis_label = "Retention Time (min)"
     p.yaxis.axis_label = "m/z"
     scatter_renderer: Any = None
-    if size.lower() in ["dyn", "dynamic"]:
+    if scaling.lower() in ["dyn", "dynamic"]:
         scatter_renderer = p.circle(
             x="rt",
             y="mz",
-            radius=markersize / 10,
+            radius=markersize,
             fill_color={"field": colorby, "transform": color_mapper},
             line_color=None,
             alpha=alpha,
@@ -414,7 +509,6 @@ def plot_samples_2d(
     size="dynamic",
     alpha_max=0.8,
     alpha="inty",
-    cmap="Turbo256",
     max_features=50000,
     width=600,
     height=600,
@@ -447,7 +541,6 @@ def plot_samples_2d(
     from bokeh.plotting import figure, show, output_file
     from bokeh.io.export import export_png
     from bokeh.models import ColumnDataSource, HoverTool
-    from bokeh.palettes import Turbo256
 
     sample_uids = self._get_sample_uids(samples)
 
@@ -455,8 +548,14 @@ def plot_samples_2d(
         self.logger.error("No valid sample_uids provided.")
         return
 
-    colors = Turbo256
-    color_map = {uid: colors[i * (256 // max(1, len(sample_uids)))] for i, uid in enumerate(sample_uids)}
+    # Get sample colors from samples_df
+    sample_colors = (
+        self.samples_df
+        .filter(pl.col("sample_uid").is_in(sample_uids))
+        .select(["sample_uid", "sample_color"])
+        .to_dict(as_series=False)
+    )
+    color_map = dict(zip(sample_colors["sample_uid"], sample_colors["sample_color"]))
 
     p = figure(
         width=width,
@@ -569,6 +668,7 @@ def plot_samples_2d(
                 "inty": sample_data["inty"].values,
                 "alpha": sample_data["alpha"].values,
                 "sample": np.full(len(sample_data), sample_name, dtype=object),
+                "sample_color": np.full(len(sample_data), color_values[uid], dtype=object),
             },
         )
 
@@ -604,6 +704,7 @@ def plot_samples_2d(
         hover = HoverTool(
             tooltips=[
                 ("sample", "@sample"),
+                ("sample_color", "$color[swatch]:sample_color"),
                 ("rt", "@rt{0.00}"),
                 ("mz", "@mz{0.0000}"),
                 ("intensity", "@inty{0.0e+0}"),
@@ -637,7 +738,6 @@ def plot_bpc(
     filename: str | None = None,
     width: int = 1000,
     height: int = 300,
-    rt_unit: str = "s",
     original: bool = False,
 ):
     """
@@ -653,7 +753,6 @@ def plot_bpc(
     from bokeh.plotting import figure, show, output_file
     from bokeh.models import ColumnDataSource, HoverTool
     from bokeh.io.export import export_png
-    from bokeh.palettes import Turbo256
     from masster.study.helpers import get_bpc
 
     sample_uids = self._get_sample_uids(samples)
@@ -664,10 +763,14 @@ def plot_bpc(
     # Debug: show which sample_uids we will process
     self.logger.debug(f"plot_bpc: sample_uids={sample_uids}")
 
-    colors = Turbo256
-    n = max(1, len(sample_uids))
-    step = max(1, 256 // n)
-    color_map = {uid: colors[(i * step) % 256] for i, uid in enumerate(sample_uids)}
+    # Get sample colors from samples_df
+    sample_colors = (
+        self.samples_df
+        .filter(pl.col("sample_uid").is_in(sample_uids))
+        .select(["sample_uid", "sample_color"])
+        .to_dict(as_series=False)
+    )
+    color_map = dict(zip(sample_colors["sample_uid"], sample_colors["sample_color"]))
 
     # If plotting original (uncorrected) RTs, use the requested title.
     if original:
@@ -675,6 +778,17 @@ def plot_bpc(
     else:
         plot_title = title or "Base Peak Chromatograms"
 
+    # Get rt_unit from the first chromatogram, default to "s" if not available
+    rt_unit = "s"
+    for uid in sample_uids:
+        try:
+            first_chrom = get_bpc(self, sample=uid, label=None, original=original)
+            if hasattr(first_chrom, 'rt_unit'):
+                rt_unit = first_chrom.rt_unit
+                break
+        except Exception:
+            continue
+
     p = figure(width=width, height=height, title=plot_title, tools="pan,wheel_zoom,box_zoom,reset,save")
     p.xaxis.axis_label = f"Retention Time ({rt_unit})"
     p.yaxis.axis_label = "Intensity"
@@ -691,7 +805,7 @@ def plot_bpc(
 
     for uid in sample_uids:
         try:
-            chrom = get_bpc(self, sample=uid, rt_unit=rt_unit, label=None, original=original)
+            chrom = get_bpc(self, sample=uid, label=None, original=original)
         except Exception as e:
             # log and skip samples we can't compute BPC for
             self.logger.debug(f"Skipping sample {uid} for BPC: {e}")
@@ -743,7 +857,7 @@ def plot_bpc(
             f"Processing BPC for sample_uid={uid}, sample_name={sample_name}, rt_len={rt.size}, color={color}"
         )
 
-        data = {"rt": rt, "inty": inty, "sample": [sample_name] * len(rt)}
+        data = {"rt": rt, "inty": inty, "sample": [sample_name] * len(rt), "sample_color": [color] * len(rt)}
         src = ColumnDataSource(data)
 
         r_line = p.line("rt", "inty", source=src, line_width=1, color=color, legend_label=str(sample_name))
@@ -754,7 +868,7 @@ def plot_bpc(
         self.logger.warning("No BPC curves to plot for the selected samples.")
         return
 
-    hover = HoverTool(tooltips=[("sample", "@sample"), ("rt", "@rt{0.00}"), ("inty", "@inty{0.0e+0}")], renderers=renderers)
+    hover = HoverTool(tooltips=[("sample", "@sample"), ("sample_color", "$color[swatch]:sample_color"), ("rt", "@rt{0.00}"), ("inty", "@inty{0.00e0}")], renderers=renderers)
     p.add_tools(hover)
 
     # Only set legend properties if a legend was actually created to avoid Bokeh warnings
@@ -784,13 +898,12 @@ def plot_bpc(
 def plot_eic(
     self,
     mz,
-    mz_tol=0.01,
+    mz_tol=None,
     samples=None,
     title: str | None = None,
     filename: str | None = None,
     width: int = 1000,
     height: int = 300,
-    rt_unit: str = "s",
     original: bool = False,
 ):
     """
@@ -799,14 +912,20 @@ def plot_eic(
     Parameters mirror `plot_bpc` with additional `mz` and `mz_tol` arguments. The function
     retrieves a Sample object for each sample UID, calls `sample.get_eic(mz, mz_tol)`, and
     overlays the resulting chromatograms.
+
+    Args:
+        mz_tol: m/z tolerance in Da. If None, uses study.parameters.eic_mz_tol as default.
     """
     # Local imports to avoid heavy top-level deps / circular imports
     from bokeh.plotting import figure, show, output_file
     from bokeh.models import ColumnDataSource, HoverTool
     from bokeh.io.export import export_png
-    from bokeh.palettes import Turbo256
     from masster.study.helpers import get_eic
 
+    # Use study's eic_mz_tol parameter as default if not provided
+    if mz_tol is None:
+        mz_tol = self.parameters.eic_mz_tol
+
     if mz is None:
         self.logger.error("mz must be provided for EIC plotting")
         return
@@ -816,13 +935,28 @@ def plot_eic(
         self.logger.error("No valid sample_uids provided for EIC plotting.")
         return
 
-    colors = Turbo256
-    n = max(1, len(sample_uids))
-    step = max(1, 256 // n)
-    color_map = {uid: colors[(i * step) % 256] for i, uid in enumerate(sample_uids)}
+    # Get sample colors from samples_df
+    sample_colors = (
+        self.samples_df
+        .filter(pl.col("sample_uid").is_in(sample_uids))
+        .select(["sample_uid", "sample_color"])
+        .to_dict(as_series=False)
+    )
+    color_map = dict(zip(sample_colors["sample_uid"], sample_colors["sample_color"]))
 
     plot_title = title or f"Extracted Ion Chromatograms (m/z={mz:.4f} ± {mz_tol})"
 
+    # Get rt_unit from the first chromatogram, default to "s" if not available
+    rt_unit = "s"
+    for uid in sample_uids:
+        try:
+            first_chrom = get_eic(self, sample=uid, mz=mz, mz_tol=mz_tol, label=None)
+            if hasattr(first_chrom, 'rt_unit'):
+                rt_unit = first_chrom.rt_unit
+                break
+        except Exception:
+            continue
+
     p = figure(width=width, height=height, title=plot_title, tools="pan,wheel_zoom,box_zoom,reset,save")
     p.xaxis.axis_label = f"Retention Time ({rt_unit})"
     p.yaxis.axis_label = "Intensity"
@@ -839,7 +973,7 @@ def plot_eic(
 
     for uid in sample_uids:
         try:
-            chrom = get_eic(self, sample=uid, mz=mz, mz_tol=mz_tol, rt_unit=rt_unit, label=None)
+            chrom = get_eic(self, sample=uid, mz=mz, mz_tol=mz_tol, label=None)
         except Exception as e:
             # log and skip samples we can't compute EIC for
             self.logger.debug(f"Skipping sample {uid} for EIC: {e}")
@@ -885,7 +1019,7 @@ def plot_eic(
 
         color = color_map.get(uid, "#000000")
 
-        data = {"rt": rt, "inty": inty, "sample": [sample_name] * len(rt)}
+        data = {"rt": rt, "inty": inty, "sample": [sample_name] * len(rt), "sample_color": [color] * len(rt)}
         src = ColumnDataSource(data)
 
         r_line = p.line("rt", "inty", source=src, line_width=1, color=color, legend_label=str(sample_name))
@@ -896,7 +1030,7 @@ def plot_eic(
         self.logger.warning("No EIC curves to plot for the selected samples.")
         return
 
-    hover = HoverTool(tooltips=[("sample", "@sample"), ("rt", "@rt{0.00}"), ("inty", "@inty{0.0e+0}")], renderers=renderers)
+    hover = HoverTool(tooltips=[("sample", "@sample"), ("sample_color", "$color[swatch]:sample_color"), ("rt", "@rt{0.00}"), ("inty", "@inty{0.0e0}")], renderers=renderers)
     p.add_tools(hover)
 
     if getattr(p, "legend", None) and len(p.legend) > 0:
@@ -928,7 +1062,6 @@ def plot_rt_correction(
     filename: str | None = None,
     width: int = 1000,
     height: int = 300,
-    rt_unit: str = "s",
 ):
     """
     Plot RT correction per sample: (rt - rt_original) vs rt overlayed for selected samples.
@@ -937,7 +1070,6 @@ def plot_rt_correction(
     """
     from bokeh.plotting import figure, show, output_file
     from bokeh.models import ColumnDataSource, HoverTool
-    from bokeh.palettes import Turbo256
     import numpy as _np
 
     # Validate features dataframe
@@ -954,11 +1086,17 @@ def plot_rt_correction(
         self.logger.error("No valid sample_uids provided for RT correction plotting.")
         return
 
-    # Color mapping like plot_bpc
-    colors = Turbo256
-    n = max(1, len(sample_uids))
-    step = max(1, 256 // n)
-    color_map = {uid: colors[(i * step) % 256] for i, uid in enumerate(sample_uids)}
+    # Get sample colors from samples_df
+    sample_colors = (
+        self.samples_df
+        .filter(pl.col("sample_uid").is_in(sample_uids))
+        .select(["sample_uid", "sample_color"])
+        .to_dict(as_series=False)
+    )
+    color_map = dict(zip(sample_colors["sample_uid"], sample_colors["sample_color"]))
+
+    # For RT correction plots, default to "s" since we're working with features_df directly
+    rt_unit = "s"
 
     p = figure(width=width, height=height, title=title or "RT correction", tools="pan,wheel_zoom,box_zoom,reset,save")
     p.xaxis.axis_label = f"Retention Time ({rt_unit})"
@@ -1026,7 +1164,7 @@ def plot_rt_correction(
 
         color = color_map.get(uid, "#000000")
 
-        data = {"rt": rt, "delta": delta, "sample": [sample_name] * len(rt)}
+        data = {"rt": rt, "delta": delta, "sample": [sample_name] * len(rt), "sample_color": [color] * len(rt)}
         src = ColumnDataSource(data)
 
         r_line = p.line("rt", "delta", source=src, line_width=1, color=color)
@@ -1037,7 +1175,7 @@ def plot_rt_correction(
         self.logger.warning("No RT correction curves to plot for the selected samples.")
         return
 
-    hover = HoverTool(tooltips=[("sample", "@sample"), ("rt", "@rt{0.00}"), ("rt - rt_original", "@delta{0.00}")], renderers=renderers)
+    hover = HoverTool(tooltips=[("sample", "@sample"), ("sample_color", "$color[swatch]:sample_color"), ("rt", "@rt{0.00}"), ("rt - rt_original", "@delta{0.00}")], renderers=renderers)
     p.add_tools(hover)
 
     # Only set legend properties if a legend was actually created to avoid Bokeh warnings
@@ -1083,15 +1221,17 @@ def plot_chrom(
         self.logger.error("No chromatogram data found.")
         return
 
-    # Local import for color palette
-    from bokeh.palettes import Turbo256
-
-    # Assign a fixed color to each sample/column
+    # Get sample colors for alignment plots
+    # Need to map sample names to colors since chromatogram data uses sample names as columns
     sample_names = [col for col in chroms.columns if col not in ["consensus_uid"]]
     if not sample_names:
         self.logger.error("No sample names found in chromatogram data.")
         return
-    color_map = {sample: Turbo256[i * (256 // max(1, len(sample_names)))] for i, sample in enumerate(sample_names)}
+    
+    # Create color mapping by getting sample_color for each sample_name
+    samples_info = self.samples_df.select(["sample_name", "sample_color"]).to_dict(as_series=False)
+    sample_name_to_color = dict(zip(samples_info["sample_name"], samples_info["sample_color"]))
+    color_map = {name: sample_name_to_color.get(name, "#1f77b4") for name in sample_names}  # fallback to blue
 
     plots = []
     self.logger.info(f"Plotting {chroms.shape[0]} chromatograms...")
@@ -1461,8 +1601,8 @@ def plot_consensus_stats(
 def plot_pca(
     self,
     filename=None,
-    width=400,
-    height=400,
+    width=500,
+    height=450,
     alpha=0.8,
     markersize=6,
     n_components=2,
@@ -1484,7 +1624,7 @@ def plot_pca(
     """
     from bokeh.models import ColumnDataSource, HoverTool, ColorBar, LinearColorMapper
     from bokeh.plotting import figure, show, output_file
-    from bokeh.palettes import Category20, viridis, Turbo256
+    from bokeh.palettes import Category20, viridis
     from bokeh.transform import factor_cmap
     from sklearn.decomposition import PCA
     from sklearn.preprocessing import StandardScaler
@@ -1507,7 +1647,7 @@ def plot_pca(
         self.logger.error("No samples dataframe available.")
         return
 
-    self.logger.info(f"Performing PCA on consensus matrix with shape: {consensus_matrix.shape}")
+    self.logger.debug(f"Performing PCA on consensus matrix with shape: {consensus_matrix.shape}")
 
     # Convert consensus matrix to numpy if it's not already
     if hasattr(consensus_matrix, "values"):
@@ -1534,7 +1674,7 @@ def plot_pca(
     # Get explained variance ratios
     explained_var = pca.explained_variance_ratio_
 
-    self.logger.info(f"PCA explained variance ratios: {explained_var}")
+    self.logger.debug(f"PCA explained variance ratios: {explained_var}")
 
     # Convert samples_df to pandas for easier manipulation
     samples_pd = samples_df.to_pandas()
@@ -1619,15 +1759,31 @@ def plot_pca(
                 legend_field=color_by,
             )
     else:
-        # If no color_by provided, color points by sample similar to plot_samples_2d
+        # If no color_by provided, use sample_color column from samples_df
         if "sample_uid" in pca_df.columns or "sample_name" in pca_df.columns:
             # Choose the identifier to map colors by
             id_col = "sample_uid" if "sample_uid" in pca_df.columns else "sample_name"
-            sample_ids = list(pd.unique(pca_df[id_col]))
-            colors = Turbo256
-            color_map = {uid: colors[i * (256 // max(1, len(sample_ids)))] for i, uid in enumerate(sample_ids)}
+            
+            # Get colors from samples_df based on the identifier
+            if id_col == "sample_uid":
+                sample_colors = (
+                    self.samples_df
+                    .filter(pl.col("sample_uid").is_in(pca_df[id_col].unique()))
+                    .select(["sample_uid", "sample_color"])
+                    .to_dict(as_series=False)
+                )
+                color_map = dict(zip(sample_colors["sample_uid"], sample_colors["sample_color"]))
+            else:  # sample_name
+                sample_colors = (
+                    self.samples_df
+                    .filter(pl.col("sample_name").is_in(pca_df[id_col].unique()))
+                    .select(["sample_name", "sample_color"])
+                    .to_dict(as_series=False)
+                )
+                color_map = dict(zip(sample_colors["sample_name"], sample_colors["sample_color"]))
+            
             # Map colors into dataframe
-            pca_df["color"] = [color_map[x] for x in pca_df[id_col]]
+            pca_df["color"] = [color_map.get(x, "#1f77b4") for x in pca_df[id_col]]  # fallback to blue
             # Update the ColumnDataSource with new color column
             source = ColumnDataSource(pca_df)
             scatter = p.scatter(
@@ -1652,14 +1808,17 @@ def plot_pca(
     tooltip_list = []
 
     # Columns to exclude from tooltips (file paths and internal/plot fields)
-    excluded_cols = {"file_source", "file_path", "sample_path", "map_id", "PC1", "PC2", "ms1", "ms2"}
+    excluded_cols = {"file_source", "file_path", "sample_path", "map_id", "PC1", "PC2", "ms1", "ms2", "size"}
 
     # Add all sample dataframe columns to tooltips, skipping excluded ones
     for col in samples_pd.columns:
         if col in excluded_cols:
             continue
         if col in pca_df.columns:
-            if pca_df[col].dtype in ["float64", "float32"]:
+            if col == "sample_color":
+                # Display sample_color as a colored swatch
+                tooltip_list.append(('color', "$color[swatch]:sample_color"))
+            elif pca_df[col].dtype in ["float64", "float32"]:
                 tooltip_list.append((col, f"@{col}{{0.00}}"))
             else:
                 tooltip_list.append((col, f"@{col}"))
@@ -1691,7 +1850,6 @@ def plot_tic(
     filename: str | None = None,
     width: int = 1000,
     height: int = 300,
-    rt_unit: str = "s",
     original: bool = False,
 ):
     """
@@ -1703,7 +1861,6 @@ def plot_tic(
     from bokeh.plotting import figure, show, output_file
     from bokeh.models import ColumnDataSource, HoverTool
     from bokeh.io.export import export_png
-    from bokeh.palettes import Turbo256
     from masster.study.helpers import get_tic
 
     sample_uids = self._get_sample_uids(samples)
@@ -1711,13 +1868,28 @@ def plot_tic(
         self.logger.error("No valid sample_uids provided for TIC plotting.")
         return
 
-    colors = Turbo256
-    n = max(1, len(sample_uids))
-    step = max(1, 256 // n)
-    color_map = {uid: colors[(i * step) % 256] for i, uid in enumerate(sample_uids)}
+    # Get sample colors from samples_df
+    sample_colors = (
+        self.samples_df
+        .filter(pl.col("sample_uid").is_in(sample_uids))
+        .select(["sample_uid", "sample_color"])
+        .to_dict(as_series=False)
+    )
+    color_map = dict(zip(sample_colors["sample_uid"], sample_colors["sample_color"]))
 
     plot_title = title or "Total Ion Chromatograms"
 
+    # Get rt_unit from the first chromatogram, default to "s" if not available
+    rt_unit = "s"
+    for uid in sample_uids:
+        try:
+            first_chrom = get_tic(self, sample=uid, label=None)
+            if hasattr(first_chrom, 'rt_unit'):
+                rt_unit = first_chrom.rt_unit
+                break
+        except Exception:
+            continue
+
     p = figure(width=width, height=height, title=plot_title, tools="pan,wheel_zoom,box_zoom,reset,save")
     p.xaxis.axis_label = f"Retention Time ({rt_unit})"
     p.yaxis.axis_label = "Intensity"
@@ -1778,7 +1950,7 @@ def plot_tic(
 
         color = color_map.get(uid, "#000000")
 
-        data = {"rt": rt, "inty": inty, "sample": [sample_name] * len(rt)}
+        data = {"rt": rt, "inty": inty, "sample": [sample_name] * len(rt), "sample_color": [color] * len(rt)}
         src = ColumnDataSource(data)
 
         r_line = p.line("rt", "inty", source=src, line_width=1, color=color, legend_label=str(sample_name))
@@ -1789,7 +1961,7 @@ def plot_tic(
         self.logger.warning("No TIC curves to plot for the selected samples.")
         return
 
-    hover = HoverTool(tooltips=[("sample", "@sample"), ("rt", "@rt{0.00}"), ("inty", "@inty{0.0e+0}")], renderers=renderers)
+    hover = HoverTool(tooltips=[("sample", "@sample"), ("sample_color", "$color[swatch]:sample_color"), ("rt", "@rt{0.00}"), ("inty", "@inty{0.00e0}")], renderers=renderers)
     p.add_tools(hover)
 
     # Only set legend properties if a legend was actually created to avoid Bokeh warnings