Merge pull request #103 from lazear/main

Add additional search engine, some fixes
Proteobench · Sep 26, 2023 · 5ce715e · 5ce715e
2 parents 584aa2d + 75aa06d
commit 5ce715e
Show file tree

Hide file tree

Showing 14 changed files with 42,835 additions and 41 deletions.
diff --git a/proteobench/modules/dda_quant/datapoint.py b/proteobench/modules/dda_quant/datapoint.py
@@ -1,4 +1,5 @@
 import json
+import numpy as np
 from dataclasses import asdict, dataclass
 from datetime import datetime
 
@@ -44,7 +45,7 @@ def calculate_plot_data(self, df):
         nr_missing_0 = 0
         for spec in species:
             f = len(df[df[spec] == True])
-            sum_s = (df[df[spec] == True]["1|2_expected_ratio_diff"]).sum()
+            sum_s = np.nan_to_num(df[df[spec] == True]["1|2_expected_ratio_diff"], nan=0, neginf=-1000, posinf=1000).sum()
             ratio = sum_s / f
             prop_ratio = (f / len(df)) * ratio
             prop_ratios.append(prop_ratio)

diff --git a/proteobench/modules/dda_quant/io_parse_settings/parse_settings_alphapept.toml b/proteobench/modules/dda_quant/io_parse_settings/parse_settings_alphapept.toml
@@ -30,10 +30,10 @@ LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_03 = "LFQ_Orbitrap_DDA_Condition_B_Sam
 
 [species_expected_ratio]
 [species_expected_ratio.YEAST]
-"1|2" = 0.5
+"1|2" = 2.0
 
 [species_expected_ratio.ECOLI]
-"1|2" = 1.5
+"1|2" = 0.25
 
 [species_expected_ratio.HUMAN]
 "1|2" = 1.0

diff --git a/proteobench/modules/dda_quant/io_parse_settings/parse_settings_maxquant.toml b/proteobench/modules/dda_quant/io_parse_settings/parse_settings_maxquant.toml
@@ -28,16 +28,16 @@ LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_03 = 2
 
 [species_expected_ratio]
 [species_expected_ratio.YEAST]
-"1|2" = 0.5
-"LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_01|LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_01" = 0.5
-"LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_02|LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_02" = 0.5
-"LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_03|LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_03" = 0.5
+"1|2" = 2.0
+"LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_01|LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_01" = 2.0
+"LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_02|LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_02" = 2.0
+"LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_03|LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_03" = 2.0
 
 [species_expected_ratio.ECOLI]
-"1|2" = 1.5
-"LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_01|LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_01" = 1.5
-"LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_02|LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_02" = 1.5
-"LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_03|LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_03" = 1.5
+"1|2" = 0.25
+"LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_01|LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_01" = 0.25
+"LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_02|LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_02" = 0.25
+"LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_03|LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_03" = 0.25
 
 [species_expected_ratio.HUMAN]
 "1|2" = 1.0

diff --git a/proteobench/modules/dda_quant/io_parse_settings/parse_settings_msfragger.toml b/proteobench/modules/dda_quant/io_parse_settings/parse_settings_msfragger.toml
@@ -28,10 +28,10 @@ Charge = "Charge"
 
 [species_expected_ratio]
 [species_expected_ratio.YEAST]
-"1|2" = 0.5
+"1|2" = 2.0
 
 [species_expected_ratio.ECOLI]
-"1|2" = 1.5
+"1|2" = 0.25
 
 [species_expected_ratio.HUMAN]
 "1|2" = 1.0

diff --git a/proteobench/modules/dda_quant/io_parse_settings/parse_settings_proline.toml b/proteobench/modules/dda_quant/io_parse_settings/parse_settings_proline.toml
@@ -28,10 +28,10 @@ abundance_DDA_Condition_B_Sample_Alpha_03 = 2
 
 [species_expected_ratio]
 [species_expected_ratio.YEAST]
-"1|2" = 0.5
+"1|2" = 2.0
 
 [species_expected_ratio.ECOLI]
-"1|2" = 1.5
+"1|2" = 0.25
 
 [species_expected_ratio.HUMAN]
 "1|2" = 1.0

diff --git a/proteobench/modules/dda_quant/io_parse_settings/parse_settings_sage.toml b/proteobench/modules/dda_quant/io_parse_settings/parse_settings_sage.toml
@@ -0,0 +1,40 @@
+[mapper]
+"proteins" = "Proteins"
+"peptide" = "Sequence"
+"charge" = "Charge"
+
+[replicate_mapper]
+"LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_01.mzML.gz" = 1
+"LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_02.mzML.gz" = 1
+"LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_03.mzML.gz" = 1
+"LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_01.mzML.gz" = 2
+"LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_02.mzML.gz" = 2
+"LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_03.mzML.gz" = 2
+
+[run_mapper]
+"LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_01.mzML" = "LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_01"
+"LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_02.mzML" = "LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_02"
+"LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_03.mzML" = "LFQ_Orbitrap_DDA_Condition_A_Sample_Alpha_03"
+"LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_01.mzML" = "LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_01"
+"LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_02.mzML" = "LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_02"
+"LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_03.mzML" = "LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_03"
+
+[species_dict]
+"YEAST" = "_YEAST"
+"ECOLI" = "_ECOLI"
+"HUMAN" = "_HUMAN"
+
+[species_expected_ratio]
+[species_expected_ratio.YEAST]
+"1|2" = 2.0
+
+[species_expected_ratio.ECOLI]
+"1|2" = 0.25
+
+[species_expected_ratio.HUMAN]
+"1|2" = 1.0
+
+[general]
+contaminant_flag = "Cont_"
+decoy_flag = false
+min_count_multispec = 1
diff --git a/proteobench/modules/dda_quant/io_parse_settings/parse_settings_wombat.toml b/proteobench/modules/dda_quant/io_parse_settings/parse_settings_wombat.toml
@@ -26,10 +26,10 @@ abundance_B_3 = "LFQ_Orbitrap_DDA_Condition_B_Sample_Alpha_03"
 
 [species_expected_ratio]
 [species_expected_ratio.YEAST]
-"1|2" = 0.5
+"1|2" = 2.0
 
 [species_expected_ratio.ECOLI]
-"1|2" = 1.5
+"1|2" = 0.25
 
 [species_expected_ratio.HUMAN]
 "1|2" = 1.0

diff --git a/proteobench/modules/dda_quant/module.py b/proteobench/modules/dda_quant/module.py
@@ -152,6 +152,8 @@ def load_input_file(self, input_csv: str, input_format: str) -> pd.DataFrame:
             input_data_frame = pd.read_csv(input_csv, sep="\t", low_memory=False)
         elif input_format == "AlphaPept":
             input_data_frame = pd.read_csv(input_csv, low_memory=False)
+        elif input_format == "Sage":
+            input_data_frame = pd.read_csv(input_csv, sep='\t', low_memory=False)
         elif input_format == "MSFragger":
             input_data_frame = pd.read_csv(input_csv, low_memory=False, sep="\t")
         elif input_format == "WOMBAT":

diff --git a/proteobench/modules/dda_quant/parse_settings.py b/proteobench/modules/dda_quant/parse_settings.py
@@ -19,6 +19,7 @@
                         "MSFragger"        : os.path.join(PARSE_SETTINGS_DIR, 'parse_settings_msfragger.toml'),
                         "Proline"         : os.path.join(PARSE_SETTINGS_DIR, 'parse_settings_proline.toml'),
                         "AlphaPept"        : os.path.join(PARSE_SETTINGS_DIR, 'parse_settings_alphapept.toml'),
+                        "Sage"        : os.path.join(PARSE_SETTINGS_DIR, 'parse_settings_sage.toml'),
                         "Custom"        : os.path.join(PARSE_SETTINGS_DIR, 'parse_settings_custom.toml')
             }
 
@@ -28,6 +29,7 @@
                 "MSFragger",
                 "Proline",
                 "WOMBAT",
+                "Sage",
                 "Custom")
 
 LOCAL_DEVELOPMENT = False

diff --git a/proteobench/modules/dda_quant/plot.py b/proteobench/modules/dda_quant/plot.py
@@ -1,7 +1,7 @@
 import numpy as np
 import pandas as pd
-import plotly.figure_factory as ff
 import plotly.graph_objects as go
+import plotly.express as px
 import streamlit as st
 from streamlit_plotly_events import plotly_events
 
@@ -11,18 +11,21 @@ class PlotDataPoint:
     def plot_bench(self, result_df: pd.DataFrame) -> go.Figure:
         """Plot results with Plotly Express."""
 
-        hist_data = [
-            np.array(result_df[result_df["YEAST"] == True]["1|2_ratio"]),
-            np.array(result_df[result_df["HUMAN"] == True]["1|2_ratio"]),
-            np.array(result_df[result_df["ECOLI"] == True]["1|2_ratio"]),
-        ]
-        group_labels = [
-            "YEAST",
-            "HUMAN",
-            "ECOLI",
-        ]
-
-        fig = ff.create_distplot(hist_data, group_labels, show_hist=False)
+        # Remove any precursors not arising from a known organism... contaminants?
+        result_df = result_df[result_df[["YEAST", "ECOLI", "HUMAN"]].any(axis=1)]
+        result_df["kind"] = result_df[["YEAST", "ECOLI", "HUMAN"]].apply(
+            lambda x: ["YEAST", "ECOLI", "HUMAN"][np.argmax(x)], axis=1
+        )
+        fig = px.histogram(
+            result_df,
+            x=np.log2(result_df["1|2_ratio"]),
+            color="kind",
+            marginal="rug",
+            histnorm="probability density",
+            barmode="overlay",
+            opacity=0.7,
+            nbins=100
+        )
 
         fig.update_layout(
             width=700,
@@ -39,9 +42,9 @@ def plot_bench(self, result_df: pd.DataFrame) -> go.Figure:
                 gridwidth=2,
             ),
         )
-        fig.update_xaxes(range=[0, 4])
+        fig.update_xaxes(range=[-4, 4])
         fig.update_xaxes(showgrid=True, gridcolor="lightgray", gridwidth=1)
-        # fig.update_yaxes(showgrid=True, gridcolor="lightgray", gridwidth=1)
+        fig.update_yaxes(showgrid=True, gridcolor="lightgray", gridwidth=1)
 
         return fig
 
@@ -66,6 +69,7 @@ def plot_metric(self, benchmark_metrics_df: pd.DataFrame) -> go.Figure:
             "MSFragger": "#ff7f0e",
             "WOMBAT": "#7f7f7f",
             "Proline": "#d62728",
+            "Sage": "#f74c00",
             "Custom": "#9467bd",
         }