NanoBoost
A Wavelet Transform-Enhanced Machine Learning Algorithm for Next-Generation Nanopore Multiplexing
NanoBoost is a machine learning pipeline combining XGBoost with a novel event-specific Discrete Wavelet Transform (DWT) for nanopore translocation event classification. It achieved 91% accuracy in nanoparticle size classification (5 nm vs 10 nm nanospheres) and 99% in shape classification (nanospheres vs nanorods).
Installation
git clone https://github.com/joehart2001/nanoboost.git
cd nanoboost
pip install -e .
For unsupervised learning and plotting dependencies:
pip install -e ".[full]"
Usage
End-to-end run
Edit config.yaml to point to your ABF files and set labels, then:
nanoboost run config.yaml
Step-by-step
# 1. Isolate events from a raw ABF recording
nanoboost preprocess recording.abf events.pkl
# 2. Apply per-event DWT and extract features (label: 5 or 10 for nm, NR or NS for shape)
nanoboost transform events.pkl features.pkl 10
# 3. Train a classifier across one or more feature sets
nanoboost train features_5nm.pkl features_10nm.pkl --output model.pkl
Use nanoboost <command> --help for full options on any command.
Configuration
config.yaml controls all parameters. Paper-optimal defaults are pre-set:
data:
files:
- path: recordings/5nm.abf
label: 5
- path: recordings/10nm.abf
label: 10
output_dir: results/
preprocessing:
resistive: false # true for biphasic events with a resistive (trough) component
transform:
wavelet: bior3.3 # paper-optimal mother wavelet
threshold: 0.2 # paper-optimal DWT coefficient threshold
training:
model: xgboost # xgboost | rf | svm | dt
search: random # random | bayes | sobol | grid
CLI flags override config values for individual commands:
nanoboost transform events.pkl features.pkl 10 --wavelet haar --threshold 0.3
nanoboost train features.pkl --model rf --search bayes
Citation
If you use this code or methodology in your research, please cite this repository (paper details to be added upon publication). Here is a suggested BibTeX entry:
@software{nanoboost,
author = {Hart, Joseph},
title = {NanoBoost: A Wavelet Transform-Enhanced Machine Learning Algorithm for Next-Generation Nanopore Multiplexing},
year = {2025},
url = {https://github.com/joehart2001/nanoboost}
}
For the complete citation information, see CITATION.cff.
License
MIT — see LICENSE.
Metadata
Release files for nanoboost 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| nanoboost-0.1.0.tar.gz | 29.4 kB | Without provenance | Details for nanoboost-0.1.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| nanoboost-0.1.0-py3-none-any.whl | Python 3 | none | any | Details for nanoboost-0.1.0-py3-none-any.whl |
Total release size: 64.7 kB
Release files / nanoboost-0.1.0.tar.gz
| Download URL | nanoboost-0.1.0.tar.gz |
|---|---|
| Size | 29.4 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
ec31065b99efce6258ab852e44a7b9c805ed825edbfca3c933ab4a0418ce7f92
|
|
BLAKE2b-256 checksum How to use checksums |
6a06b4a765c55243bd81603def51369b1ff4e23b5cdf1ce6be49d151b407d93b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.8.18
|
Release files / nanoboost-0.1.0-py3-none-any.whl
| Download URL | nanoboost-0.1.0-py3-none-any.whl |
|---|---|
| Size | 35.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
4a47d6e4e3052e84ef98a2fb50be50cf63c8e5fc68fd481f89f80d23e2f74c6f
|
|
BLAKE2b-256 checksum How to use checksums |
dc55a5af5d1526d1766fcc34323e58741866511b1e2f61a4241ea642e3d7963c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.8.18
|
Security for nanoboost 0.1.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.