yuxuanliao / dffroi

Fusion of Quality Evaluation Metrics and Convolutional Neural Network Representations for Accurate Peak Detection in LC-MS

1.Introduction

The region of interests (ROIs) extraction is the most fundamental step in analyzing metabolomic dataset acquired by liquid chromatography mass spectrometry (LC-MS). However, noises and backgrounds existing in LC-MS data often affect the quality of extracted ROIs. Therefore, the development of effective ROIs evaluation algorithms is necessary to eliminate the false positives meanwhile keep the false negative rate as low as possible. In this study, deep fused filter of ROIs (dffROI) was proposed to improve the accuracy of ROI extraction by combing the handcrafted evaluation metrics with convolution-al neural network (CNN)-learned representations. Results show that dffROI can achieve higher accuracy, sensitivity and lower false positive rate. The model-agnostic feature importance demon-strates the necessity of fusing handcrafted evaluation metrics with the convolutional neural network representations. DffROI is an automatic, robust and universal method for ROI filtering by virtue of information fusion and end-to-end learn-ing. It has been integrated into KPIC2 framework previously proposed by our group to facilitate real metab-olomic LC-MS dataset analysis.

2.Depends

Anaconda for python 3.6
TensorFlow 2.4.0
R 4.0.2

3.Install

1. Install Anaconda.
2. Install Git.
3. Install R 4.0.2.
4. Install KPIC2 in R language. The method of installing KPIC2 can refer to https://github.com/hcji/KPIC2. First install the depends of KPIC2.

install.packages(c("BiocManager", "devtools", "Ckmeans.1d.dp", "Rcpp", "RcppArmadillo", "mzR", "parallel", "shiny", "plotly", "data.table", "GA", "IRanges", "dbscan", "randomForest"))
BiocManager::install(c("mzR","ropls"))

Then, download the source package of KPIC2 at url and install the package locally.
5. Open commond line, create environment and enter with the following commands.

conda create -n dffROI python=3.6
conda activate dffROI

6. Clone the repository and enter.

git clone https://github.com/zhanghailiangcsu/dffROI.git
cd dffROI

7. Install dependency with the following commands.

pip install -r requirements.txt

8. Set environment variables for calling R language using rpy2. R_HOME represents the installation location of the R language. R_USER represents the installation location of the rpy2 package.

setx "R_HOME" "C:\Program Files\R\R-4.0.2"
setx "R_USER" "C:\Users\Administrator\Anaconda3\Lib\site-packages\rpy2"

9. Run example.py. Then, you can use dffROI+KPIC2 to process your data

python example.py

4.Usage

The dffROI is public at homepage, every user can download and use it. All ROIs can be input into dffROI for processing to filter false positives in the ROIs.
We provide an example of processing real samples using dffROI+KPIC2. It is named example(.ipynb) and we upload it. User can refer to it for processing data. The training example data in (.ipynb) is just an example to illustrate the training process. If you want to get the complete data, you can contact the author of peakonly.

5.Contact

Hailiang Zhang
E-mail [email protected]