From ac0134505a3a5530d36aeb01ba221e3c4352a302 Mon Sep 17 00:00:00 2001
From: Schneider Leo <leo.schneider@etu.ec-lyon.fr>
Date: Fri, 20 Sep 2024 17:25:49 +0200
Subject: [PATCH] fix vocab + update local integration
---
common_dataset.py | 2 +-
data_viz.py | 6 ++--
local_integration_msms.py | 70 ++++++++++++++++++++++++++++++++-------
mzml_exploration.py | 2 +-
4 files changed, 63 insertions(+), 17 deletions(-)
diff --git a/common_dataset.py b/common_dataset.py
index 6e46845..5ab8a5f 100644
--- a/common_dataset.py
+++ b/common_dataset.py
@@ -95,7 +95,7 @@ def alphabetical_to_numerical(seq, vocab):
else :
for i in range(len(seq) - 2 * seq.count('-')):
if seq[i + dec] != '-':
- num.append(IUPAC_VOCAB[seq[i + dec]])
+ num.append(ALPHABET_UNMOD[seq[i + dec]])
else:
if seq[i + dec + 1:i + dec + 4] == 'CaC':
num.append(21)
diff --git a/data_viz.py b/data_viz.py
index 23dec3e..ba16bc0 100644
--- a/data_viz.py
+++ b/data_viz.py
@@ -229,9 +229,9 @@ def add_length(dataframe):
dataframe['length']=dataframe['seq'].map(fonc)
-df = pd.read_csv('output/out_ISA_no_tape.csv')
+df = pd.read_csv('output/out_prosit_common.csv')
add_length(df)
df['abs_error'] = np.abs(df['rt pred']-df['true rt'])
-histo_abs_error(df, display=False, save=True, path='temp.png')
+# histo_abs_error(df, display=False, save=True, path='temp.png')
# scatter_rt(df, display=False, save=True, path='temp.png')
-# histo_length_by_error(df, 10, save=True, path='temp.png')
\ No newline at end of file
+histo_length_by_error(df, 10, save=True, path='temp.png')
\ No newline at end of file
diff --git a/local_integration_msms.py b/local_integration_msms.py
index 7cc384b..984e4c2 100644
--- a/local_integration_msms.py
+++ b/local_integration_msms.py
@@ -1,6 +1,7 @@
import pyopenms as oms
import numpy as np
import matplotlib.pyplot as plt
+import pandas as pd
def compute_chromatograms(rt, mz, intensity, start_c, end_c):
value=[]
@@ -12,29 +13,74 @@ def compute_chromatograms(rt, mz, intensity, start_c, end_c):
return value
+def get_df(expe, long: bool = False):
+ """Generates a pandas DataFrame with all peaks in the MSExperiment
-if __name__ == "__main__":
- e = oms.MSExperiment()
- oms.MzMLFile().load("data/Staph140.mzML", e)
- e.updateRanges()
+ Parameters:
+ long: set to True if you want to have a long/expanded/melted dataframe with one row per peak. Faster but
+ replicated RT information. If False, returns rows in the style: rt, _np.array(mz), _np.array(int)
+
+ Returns:
+ pandas.DataFrame: feature information stored in a DataFrame
+ """
+ if long:
+ cols = ["RT", "mz", "inty", 'MSlevel']
+ expe.updateRanges()
+ spectraarrs2d = expe.get2DPeakDataLong(expe.getMinRT(), expe.getMaxRT(), expe.getMinMZ(), expe.getMaxMZ())
+ return pd.DataFrame(dict(zip(cols, spectraarrs2d))) #TODO ajouter MSlevel
+
+ cols = ["RT", "mzarray", "intarray", 'MSlevel','MS1 MZ']
+
+ return pd.DataFrame(data=((spec.getRT(), *spec.get_peaks(), spec.getMSLevel(), spec.getPrecursors()[0].getMZ() if spec.getMSLevel() ==2 else None) for spec in expe), columns=cols)
+
+def generate_RT_int_imgs(exp,star_mz,stop_mz):
+ exp.updateRanges()
rt = []
- charge = []
+ mz = []
intensity = []
- for s in e :
+ for s in exp :
if s.getMSLevel() == 1:
rt.append(s.getRT())
- charge.append(s.get_peaks()[0])
+ mz.append(s.get_peaks()[0])
intensity.append(s.get_peaks()[1])
- mz_range = np.linspace(350,1250,4000)
+ mz_range = np.linspace(star_mz,stop_mz,1000)
for i in range(len(mz_range)-1):
- print(mz_range[i],'/1250')
- val = compute_chromatograms(rt, charge, intensity, mz_range[i] ,mz_range[i+1])
+ print(mz_range[i],'/{}'.format(stop_mz))
+ val = compute_chromatograms(rt, mz, intensity, mz_range[i] ,mz_range[i+1])
fig, ax = plt.subplots()
- ax.plot(val)
+ ax.plot(rt,val)
ax.set_xlabel('Retention time')
ax.set_ylabel('Intensity')
ax.set_title('mz : {} to {}'.format(mz_range[i] ,mz_range[i+1]))
plt.savefig('fig/rt_local/{}_to_{}.png'.format(mz_range[i] ,mz_range[i+1]))
plt.clf()
-df = e.get_df()
+
+def integrate_ms_ms(time_start, time_end, df):
+ df_useful = df[(df['MS1 RT']>time_start) & (df['MS1 RT']<time_end) & (df['MSlevel']==2)].reset_index(inplace=True)
+
+
+
+
+ return value
+
+if __name__ == "__main__":
+ e = oms.MSExperiment()
+ oms.MzMLFile().load("data/Staph140.mzML", e)
+ # generate_RT_int_imgs(e, 350, 1250)
+
+ df = get_df(e)
+ df1 = df[df['MSlevel'] == 1]
+ df1.reset_index(inplace=True, drop=True)
+ for i in range(len(df1)):
+ fig, ax = plt.subplots()
+ ax.plot(df1['mzarray'][i], df1['intarray'][i],linewidth=0.1)
+ ax.set_xlabel('mz')
+ ax.set_xlim(350,750)
+ ax.set_ylabel('Intensity')
+ ax.set_title('RT : {}'.format(df1['RT'][i]))
+ plt.savefig('fig/rt_local/RT{}.png'.format(df1['RT'][i]))
+ plt.close()
+
+
+
#358.1 358.32
\ No newline at end of file
diff --git a/mzml_exploration.py b/mzml_exploration.py
index 7e63464..15d80c2 100644
--- a/mzml_exploration.py
+++ b/mzml_exploration.py
@@ -301,5 +301,5 @@ if __name__ == "__main__":
# b = [np.pad(array, (0, max_len - len(array)), mode='constant', constant_values=default_value) for array in res]
-
+s = oms.MSSpectrum
--
GitLab