Skip to content

Latest commit

 

History

81 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 

Repository files navigation

MutaPheno

About

MutaPheno is an interpretable machine learning tool designed to predict both pathogenic and cancer driver missense mutations. It integrates 34 molecular features across three biological levels: residue, mutation, and protein, spanning structural, functional, physicochemical, and contextual dimensions. Using an optimized random forest classifier trained through extensive grid search and cross-validation, MutaPheno delivers robust and generalizable performance. By quantifying feature pattern enrichment between pathogenic and benign mutations via log₁₀ odds ratios, the tool achieves both high interpretability and strong predictive power. The tool accepts a list of missense mutations with corresponding protein identifiers and outputs: Predicted probabilities and binary classifications; Detailed feature annotations.

MutaPheno Installation and Usage Instructions

1. Pull Docker Image

Ensure Docker is installed:
Docker Installation Guide (https://docs.docker.com/get-started/get-docker/)

docker pull minghuilab/mutapheno:v1

• Image size: 57.3 GB
• Estimated download time (approximate):

100 Mbps: ~76 min
1 Gbps: ~8 min
(actual time depends on your network speed)

2. Run the Docker Container and Activate Environment

docker run -it minghuilab/mutapheno:v1 /bin/bash
conda activate py38
cd mutapheno

3. Preparing Input Files

Inside the Docker container, create a folder to store input files:

mkdir -p /mutapheno/input_folder

Alternatively, use the predefined directory: /mutapheno/mutapheno_input for input file storage.

Prepare input mutations in one of the following formats:

(1) UniProt (default):

Input example (example_uniprot.csv):

P21359-2|R1276P
P04637|H178Q
P27986|G376R
P04637|I251S

Command:

python MutaPheno.py \
  --input /mutapheno/mutapheno_input/example_uniprot.csv \
  --input_mutatype UniProt \
  --output /mutapheno/mutapheno_output

(2) Ensembl (Transcript/Protein):

Input example (example_ensembl.csv):

ENST00000227163|P127T
ENST00000334192|A287T
ENSP00000366506|V39I

Command:

python MutaPheno.py \
  --input /mutapheno/mutapheno_input/example_ensembl.csv \
  --input_mutatype Ensembl \
  --output /mutapheno/mutapheno_output

(3) RefSeq (Transcript/Protein):

Input example (example_refseq.csv):

NM_000267|R1276P
NP_722540|D18T

Command:

python MutaPheno.py \
  --input /mutapheno/mutapheno_input/example_refseq.csv \
  --input_mutatype RefSeq \
  --output /mutapheno/mutapheno_output

(4) Gene Name:

Input example (example_gene.csv):

BRAF|Y472C
BRCA1|C24R
EGFR|D837A
POLE|P172S

Command:

python MutaPheno.py \
  --input /mutapheno/mutapheno_input/example_gene.csv \
  --input_mutatype GeneName \
  --output /mutapheno/mutapheno_output

Important Notes:

Do not mix formats. Verify mutation identifiers, residue positions, and amino acid changes. Only validated missense mutations proceed to annotation and prediction.

4. Understanding Output Files

After running MutaPheno, three types of output files will be generated in the specified output directory:

(1) Feature Annotation (*_MutaPheno_FeatureAnnotations.csv):

This file contains the raw feature annotation data for each mutation.

(2) Prediction Results (*_MutaPheno_Predictions.csv):
Example content:

Original_Mutation	Canonical_Mutation	MutaPheno_score	MutaPheno_label
P21359-2|R1276P	P21359_R1276P	0.593913043	1
P04637|H178Q	P04637_H178Q	0.78	1
P27986|G376R	P27986_G376R	0.740869565	1
P04637|I251S	P04637_I251S	0.985652174	1

Original_Mutation: The mutation identifier as provided in the input file, including isoform-specific or transcript-specific annotations (e.g., P21359-2|R1276P).
Canonical_Mutation: The standardized mutation mapped to the canonical protein sequence (e.g., P21359_R1276P).
MutaPheno_score: Estimated probability of pathogenicity (range 0–1).
MutaPheno_label: Binary classification of pathogenicity, determined using a threshold of 0.5.

(3) Feature Contribution (*_MutaPheno_FeatureSHAP.csv):

This file contains SHAP (SHapley Additive exPlanations) values for each mutation. SHAP values quantify the contribution of individual features to the model’s prediction. A positive SHAP value indicates that the feature increases the likelihood of pathogenicity, whereas a negative value suggests the opposite.

5. Recommended System Requirements

RAM: ≥32 GB recommended (≥16 GB minimal requirement)

Disk Space: ≥100 GB (Docker image: ~57.3 GB + additional space for data and output)

About

MutaPheno: An Interpretable Molecular Phenotype-Based Classifier for Predicting Pathogenic and Cancer Driver Missense Mutations

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors