Update Text2Model

README.md

@@ -9,7 +9,7 @@ Workflow for classifying breast cancer subtypes based on intracellular signaling
 
 | Language      | Dependent packages                                             |
 | ------------- | -------------------------------------------------------------- |
-| Python >= 3.7 | [pasmopy](https://github.com/pasmopy/pasmopy)                  |
+| Python >= 3.7 | See [`requirements.txt`](requirements.txt)                     |
 | Julia >= 1.5  | [BioMASS.jl](https://github.com/himoto/BioMASS.jl)             |
 | R >= 4.0      | TCGAbiolinks, sva, biomaRt, edgeR, ComplexHeatmap, viridisLite |
 
@@ -41,7 +41,7 @@ Workflow for classifying breast cancer subtypes based on intracellular signaling
    ```python
    from pasmopy import Text2Model
 
-   Text2Model("models/erbb_network.txt").to_biomass_model()
+   Text2Model("models/erbb_network.txt").convert()
    ```
 
 1. Rename `erbb_network/` to CCLE_name or TCGA_ID, e.g., `MCF7_BREAST` or `TCGA_3C_AALK_01A`
@@ -125,7 +125,7 @@ Workflow for classifying breast cancer subtypes based on intracellular signaling
    ```python
    from pasmopy import Text2Model
 
-   Text2Model("models/erbb_network.txt", lang="julia").to_biomass_model()
+   Text2Model("models/erbb_network.txt", lang="julia").convert()
    ```
 
 1. Add time-series data to [`experimental_data.jl`](training/erbb_network_jl/experimental_data.jl)
@@ -226,4 +226,4 @@ Workflow for classifying breast cancer subtypes based on intracellular signaling
 
 ## License
 
-[Apache License 2.0](https://github.com/pasmopy/breast_cancer/blob/master/LICENSE)
+[Apache License 2.0](LICENSE)

models/breast/TCGA_3C_AALK_01A/viz.py

@@ -193,7 +193,7 @@ def set_param_range_rcParams():
     @staticmethod
     def set_sensitivity_rcParams():
         """figure/sensitivity"""
-        plt.rcParams["font.size"] = 12
+        plt.rcParams["font.size"] = 20
         plt.rcParams["axes.linewidth"] = 1.2
         plt.rcParams["xtick.major.width"] = 1.2
         plt.rcParams["ytick.major.width"] = 1.2

models/erbb_network.txt

@@ -321,9 +321,9 @@ cMycPP is degraded
 # observables
 # p: parameters
 # u: species
-@obs Phosphorylated_Akt = u[AktP]
-@obs Phosphorylated_ERK = u[ERKP] + u[ERKPP]
-@obs Phosphorylated_c-Myc = u[cMycP] + u[cMycPP]
+@obs Phosphorylated_Akt: u[AktP]
+@obs Phosphorylated_ERK: u[ERKP] + u[ERKPP]
+@obs Phosphorylated_c-Myc: u[cMycP] + u[cMycPP]
 
 @sim tspan: [0, 120]
 @sim condition EGF: init[EGF] = 10.0; init[HRG] = 0.0

training/README.md

@@ -0,0 +1,24 @@
+## Parameter estimation
+
+Using experimental datasets obtained from breast cancer cell lines to identify model parameters.
+
+### Requirements
+
+| Language                            | Package                                              | Desctiption                                                                                                           |
+| ----------------------------------- | ---------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------- |
+| [Julia 1.5+](https://julialang.org) | [`BioMASS.jl`](https://github.com/himoto/BioMASS.jl) | This module provides a Julia interface to the [biomass](https://github.com/okadalabipr/biomass) parameter estimation. |
+
+### Run optimization
+
+```bash
+$ mkdir errout
+$ sh optimize_parallel.sh
+```
+
+### How to track optimization process
+
+The temporary result will be saved in [`erbb_network_jl/logs/n.log`](https://github.com/pasmopy/breast_cancer/tree/master/training/erbb_network_jl/logs) after each iteration.
+
+```bash
+$ tail erbb_network_jl/logs/1.log
+```

training/erbb_network_jl/load_csv.jl

@@ -1,7 +1,7 @@
 using DataFrames, CSV
 
 
-const postComBat = CSV.read("./TPM_RLE_postComBat.csv", DataFrame);
+const postComBat = CSV.read("../transcriptomic_data/TPM_RLE_postComBat.csv", DataFrame);
 
 
 function get_tpm(gene::String, cell_line::String)::Float64
@@ -111,11 +111,11 @@ function incorporate_RSK(p::Vector{Float64}, cell_line::String)::Float64
             p[C.w_RPS6KA6] * get_tpm("RPS6KA6", cell_line))
 end =#
 
-#=
+#= 
 function incorporate_PP2A(p::Vector{Float64}, cell_line::String)::Float64
     return (p[C.w_PPP2CA] * get_tpm("PPP2CA", cell_line) +
             p[C.w_PPP2CB] * get_tpm("PPP2CB", cell_line))
-end=#
+end =#
 
 function incorporate_PTP1B(p::Vector{Float64}, cell_line::String)::Float64
     return p[C.w_PTPN1] * get_tpm("PTPN1", cell_line)

training/erbb_network_jl/set_search_param.jl

@@ -1,3 +1,5 @@
+include("./load_csv.jl")
+
 # Specify model parameters and/or initial values to optimize
 function get_search_index()::Tuple{Array{Int64,1},Array{Int64,1}}
     # parameters