/lingress

The lingress project is to develop the pipeline to analyse the Nuclear magnetic resonance (NMR) dataset using a univariate linear regression model. This package contains the analysis with linear regression (OLS) and visualises the interpretation of the results with a p-value of all NMR peaks.

Primary LanguagePythonMIT LicenseMIT

lingress

The Lingress project is an initiative aimed at developing a streamlined pipeline for the analysis of Nuclear Magnetic Resonance (NMR) datasets, utilizing a univariate linear regression model. This package encompasses the execution of linear regression analysis via the Ordinary Least Squares (OLS) method and provides visual interpretations of the resultant data. Notably, it includes the p-values of all NMR peaks in its analytical scope.

Functionally, this program strives to fit a model of metabolic profiles through the application of linear regression. Its design and capabilities present a robust tool for in-depth and nuanced data analysis in the realm of metabolic studies.

How to install

pip install lingress

UI Peak Picking

#Example data
import numpy as np
from lingress import pickie_peak
import pandas as pd


df = pd.read_csv("https://raw.githubusercontent.com/aeiwz/example_data/main/dataset/Example_NMR_data.csv")
spectra = df.iloc[:,1:]
ppm = spectra.columns.astype(float).to_list()


#defind plot data and run UI
pickie_peak(spectra=spectra, ppm=ppm).run_ui()

img1

Linear Regression model

import pandas as pd
from lingress import lin_regression


df = pd.read_csv("https://raw.githubusercontent.com/aeiwz/example_data/main/dataset/Example_NMR_data.csv")
X = df.iloc[:,1:]
ppm = spectra.columns.astype(float).to_list()
y = df['Group']


mod = lin_regression(x=X, target=y, label=y, features_name=ppm, adj_method='fdr_bh')
mod.create_dataset()
mod.fit_model()
mod.spec_uniplot()

spec uniplot

mod.volcano_plot()

volcano

mod.resampling(n_jobs=-1, n_boots=100, adj_method='fdr_bh')
[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.
[Parallel(n_jobs=-1)]: Done   6 tasks      | elapsed:    3.7s
[Parallel(n_jobs=-1)]: Done  60 tasks      | elapsed:    6.7s
[Parallel(n_jobs=-1)]: Done 150 tasks      | elapsed:   11.2s
[Parallel(n_jobs=-1)]: Done 276 tasks      | elapsed:   17.8s
...
[Parallel(n_jobs=-1)]: Done 6486 tasks      | elapsed:  5.6min
[Parallel(n_jobs=-1)]: Done 7188 tasks      | elapsed:  6.1min
[Parallel(n_jobs=-1)]: Done 7211 out of 7211 | elapsed:  6.1min finished

mod.resampling_df()
P-value std P-value Beta coefficient std Beta Mean P-value (F-test) std P-value (F-test) Mean R-square std R-square R2 std R-square Adjustment q_value
0.60075 3.575454e-03 1.610523e-02 3.673194e+06 502596.020205 0.434302 0.276809 0.138650 0.156244 0.030981 4.012856e-03
0.60125 2.327687e-04 6.418472e-04 4.208365e+06 638734.119190 NaN NaN 0.160225 0.175463 0.056503 3.531443e-04
0.60175 1.511846e-04 3.690541e-04 4.776924e+06 582175.023885 0.272894 0.258094 0.250765 0.204542 0.157111 2.443829e-04
0.60225 2.724337e-04 7.138873e-04 4.450884e+06 624407.676115 0.132108 0.188570 0.379931 0.198055 0.302422 4.037237e-04
0.60275 2.271675e-04 5.238926e-04 3.596622e+06 643161.588649 0.030732 0.056968 0.558447 0.158948 0.503253 3.458106e-04
... ... ... ... ... ... ... ... ... ... ...
4.20375 2.542707e-09 1.077483e-08 2.231841e+07 479783.299949 NaN NaN 0.099255 0.130321 -0.010838 4.472063e-08
4.20425 4.727199e-10 1.269310e-09 2.201865e+07 631164.491894 0.420162 0.308196 0.163733 0.184153 0.059199 1.940690e-08
4.20475 1.710447e-09 4.659603e-09 2.285026e+07 721568.566334 NaN NaN 0.100927 0.138527 -0.010207 3.595928e-08
4.20525 1.043658e-08 9.454456e-08 2.449345e+07 287615.593479 0.310386 0.301403 0.263740 0.245996 0.171707 1.084412e-07
4.20575 1.606948e-08 1.123188e-07 2.621135e+07 246414.620688 0.242344 0.257300 0.299881 0.244772 0.212366 1.457572e-07