#nJSD

nJSD is code for calculating distance between two biological networks instantiated with gene-expression profiles using entropy concept. It was designed to measure intratumor heterogeneity from bulk RNA-sequencing data. Transcriptome-based ITH (tITH) of tumor state was calculated by considering both normal state and ideally heterogeneous state.

Requirements

• Linux/Unix
• Python 2.7
• NumPy
• NetworkX

Simple nJSD

It compute distance (nJSD) between two GEPs and calculate tITH score.

run_simple.py [-h] [-n NETWORK] [-r R_GEP] [-q Q_GEP]
-h, --help  show this help message and exit
-n NETWORK  Location to network file: geneA geneB
-r R_GEP    File name of Refernece gene-expression profile
-q Q_GEP    File name of Query gene-expression profile

Network file must follow below format.

GeneA GeneB               # Header
GeneSymbol1 GeneSymbol2
GeneSymbol1 GeneSymbol3
GeneSymbol1 GeneSymbol4
...

GEP file must follow below format.

GeneSymbol  ExpressionValue       # Header
GeneA 10
GeneB 20
BeneC 30
....

In the example directory, there are test data.

example:

./run_simple.py -n example/Toy.network -r example/Toy.profile1 -q example/Toy.profile2

result:

example/Toy.profile2    [Ref. -> Query: 0.003935]       [Query -> stateH: 0.007693]     <tITH: 0.338413>

Citation

tba