Everest

Peaks identification using CNN

Usage

To install everest, download the release and run the following pip command:

pip install everest-0.0.1-py3-none-any.whl

Import the library into your python project to use everest:

from everest import find_peaks

peaks_loc, peaks_mag, probs = find_peaks(X, threshold=0.5)

Parameters

• X (array-like): Input array representing the signal.
• threshold (float): Probability threshold for identifying peaks (default=0.5).
• return_probs (bool): Set this to True to return probs (default=True).

Returns

• peaks_loc (numpy.array): Locations of the peaks.
• peaks_mag (numpy.array): Values of the peaks.
• probs (numpy.array): Probabilities associated with each peak.

Model

graph LR;
    Conv1D1[<b>Conv</b> \n<span style="font-size: smaller;">20 channels</span>] --> MaxPool[MaxPool];
    MaxPool --> Conv1D2[Conv1D\n<span style="font-size: smaller;">40 channels</span>];
    Conv1D2 --> Flatten[Flatten];
    Flatten --> Dense1[FC \n<span style="font-size: smaller;">20 units</span>];
    Dense1 --> Dense2[FC \n<span style="font-size: smaller;">4 units</span>];
    Dense2 --> Output[$$\sigma$$];

    class Conv1D1,MaxPool,Conv1D2,Flatten,Dense1,Dense2,Output non-interactive;

Loading

Results

Contributing

Contributions to Everest are welcome! If you'd like to contribute, follow these steps:

1. Fork the Repository: Start by forking the Everest.
2. Make Changes: Create a new branch , make your changes, and commit them to your branch.
3. Create a Pull Request: Push your changes to your fork and submit a pull request to the original repository.

License

This project is licensed under the GNU General Public License. See the LICENSE for details.

References

• Anne Bech Risum, Rasmus Bro, Using deep learning to evaluate peaks in chromatographic data, 2019
• Arsenty D. Melnikov, Yuri P. Tsentalovich, Vadim V. Yanshole, Deep Learning for the Precise Peak Detection in High-Resolution LC-MS Data, 2019