Emergenet

Directory Structure

The file tree shows relavent directories for the current version of the project.

Emergenet
├── emergenet : Emergenet package source code
├── examples : examples using the emergenet.emergenet and emergenet.domseq modules
├── paper_data_v3 : results for current version of the paper
└── tex : LaTeX source files for paper

Description

• Computing and predicting dominant viral strains
• Superfast risk assessment of emerging pathogens

Installation

PyPI

To install with pip:

pip install emergenet

# to update
pip install emergenet --upgrade

Dependencies

• quasinet
• numpy
• pandas
• matplotlib
• distance
• biopython
• scikit-learn
• shapely
• alphashape

Usage

Examples are located here.

Predicting Dominant Sequences

from emergenet.domseq import DomSeq, save_model, load_model

# initialize DomSeq
domseq = DomSeq(seq_trunc_length=565, random_state=42)

# load data from current time period
df = domseq.load_data(filepath='sequences.fasta')

# compute dataframe of cluster-wise dominant sequences for current time period
dominant_sequences = domseq.compute_domseq(seq_df=df)

# train enet
enet_model = domseq.train(seq_df=df, sample_size=3000)
save_model(enet=enet_model, outfile='enet_modes/')

# load candidate sequences for recommendation
pred_df = domseq.load_data(filepath='pred_sequences.fasta')

# compute dataframe of cluster-wise recommendation sequences for next time period
prediction_sequences = domseq.predict_domseq(seq_df=df, pred_seq_df=pred_df, enet=enet_model, sample_size=3000)

Evaluating Sequence Risk

from emergenet.emergenet import Enet, predict_irat_emergence

# Initialize the Enet
enet = Enet(analysis_date, ha_seq, na_seq, save_data=SAVE_DIR, random_state=42)

# Estimate the Enet risk scores
ha_risk, na_risk = enet.risk(sample_size=10000)

# Map the Enet risk scores to the IRAT risk scale
irat, irat_low, irat_high = predict_irat_emergence(ha_risk, na_risk)

Documentation

For more documentation, see here.

For a quick start guide, see here.