Calculating line widths for flux maps

For this, rather than taking values from solutions the values are inputted manually from the escher maps. It would be better from the solution dfs but I don’t fancy coding this at the moment…

import itertools

import numpy as np
import pandas as pd

conditions_permutations = [
    ["blue", "white", "nops"],
    ["unconstrained", "constrained"],
    ["wt", "starchko"],
    ["fluxes", "minimum", "maximum"],
]

solutions_df_columns = pd.MultiIndex.from_product(
    conditions_permutations, names=["Light", "ATPase", "Starch", "Solution"]
)
solutions_df = pd.DataFrame(columns=solutions_df_columns)

conditions_list_of_lists = [
    ["blue", "white", "nops"],
    ["unconstrained", "constrained"],
    ["wt", "starchko"],
]
conditions_iterations = list(itertools.product(*conditions_list_of_lists))

cols_to_drop = []
for conditions in conditions_iterations:
    light_condition, atpase_condition, starch_condition = conditions
    file_name = f"../outputs/model_solutions/{light_condition}_{atpase_condition}_{starch_condition}.csv"
    try:
        condition_df = pd.read_csv(file_name, index_col=0)
        for col in ["fluxes", "minimum", "maximum"]:
            solutions_df.loc[
                :, (light_condition, atpase_condition, starch_condition, col)
            ] = condition_df.loc[:, col]
    except:
        cols_to_drop.append(conditions)
        print(f"File {file_name} doesn't exist, skipping")


solutions_df_unphased = solutions_df.copy()
index_phased = pd.MultiIndex.from_tuples(
    [
        (reaction[:-2], int(reaction[-1]))
        if reaction[-1] in ["1", "2", "3", "4"]
        else (reaction, None)
        for reaction in solutions_df.index
    ],
    names=["Reaction", "Phase"],
)
solutions_df.index = index_phased
solutions_df = solutions_df.drop(cols_to_drop, axis=1)

/tmp/ipykernel_4834/2276567288.py:27: DeprecationWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`
  solutions_df.loc[
/tmp/ipykernel_4834/2276567288.py:27: DeprecationWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`
  solutions_df.loc[
/tmp/ipykernel_4834/2276567288.py:27: DeprecationWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`
  solutions_df.loc[
/tmp/ipykernel_4834/2276567288.py:27: DeprecationWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`
  solutions_df.loc[
/tmp/ipykernel_4834/2276567288.py:27: DeprecationWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`
  solutions_df.loc[
/tmp/ipykernel_4834/2276567288.py:27: DeprecationWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`
  solutions_df.loc[
/tmp/ipykernel_4834/2276567288.py:27: DeprecationWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`
  solutions_df.loc[
/tmp/ipykernel_4834/2276567288.py:27: DeprecationWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`
  solutions_df.loc[
/tmp/ipykernel_4834/2276567288.py:27: DeprecationWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`
  solutions_df.loc[
/tmp/ipykernel_4834/2276567288.py:27: DeprecationWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`
  solutions_df.loc[
/tmp/ipykernel_4834/2276567288.py:27: DeprecationWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`
  solutions_df.loc[
/tmp/ipykernel_4834/2276567288.py:27: DeprecationWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`
  solutions_df.loc[

solutions_df = solutions_df.xs("fluxes", level="Solution", axis=1).xs(2, level="Phase")

solutions_df

Light	blue				white				nops
ATPase	unconstrained		constrained		unconstrained		constrained		unconstrained		constrained
Starch	wt	starchko	wt	starchko	wt	starchko	wt	starchko	wt	starchko	wt	starchko
Reaction
EX_X_pi_t_me	0.000000	0.000000e+00	0.000000e+00	0.000000	0.000000e+00	0.000000	0.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000
EX_X_pi_t_gc	0.000000	0.000000e+00	0.000000e+00	0.000000	0.000000e+00	0.000000	0.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000
EX_X_CO2_t_me	4.128357	4.128357e+00	4.128356e+00	4.128355	-5.388402e+01	-53.884016	-53.884016	-5.388402e+01	-53.884046	-53.884046	-53.884048	-53.884048
EX_X_CO2_t_gc	0.003663	4.036220e-03	2.176874e-03	0.001631	1.156296e-03	0.001168	-0.001087	-1.371060e-03	0.002922	0.002922	0.001837	0.002173
EX_X_Mg_t_me	0.000000	0.000000e+00	0.000000e+00	0.000000	0.000000e+00	0.000000	0.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000
...	...	...	...	...	...	...	...	...	...	...	...	...
MAL_total_pseudolinker	0.004445	4.525811e-03	2.358282e-03	0.001766	4.455754e-03	0.004487	0.000000	-6.494415e-10	0.002314	0.002314	0.000918	0.002355
FRU_total_pseudolinker	0.000000	0.000000e+00	1.737018e-03	0.006863	0.000000e+00	0.000000	0.001898	6.805909e-03	0.000347	0.000347	0.004141	0.006863
K_total_pseudolinker	0.013250	1.303211e-02	9.057052e-03	0.005820	1.292071e-02	0.012970	0.005393	4.495170e-03	0.009849	0.009849	0.006187	0.005980
Cl_total_pseudolinker	0.011572	1.303197e-02	7.424397e-03	0.004597	1.292071e-02	0.012970	0.005393	4.495170e-03	0.008247	0.008247	0.004350	0.004350
CIT_total_pseudolinker	0.000000	4.690523e-08	-4.632768e-10	0.000000	-2.844564e-18	0.000000	0.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000

1774 rows × 12 columns

reaction_dict = {
    "starch_to_glc": "RXN_1827_p_gc_2",
    "starch_to_mal": "MALTODEG_RXN_c_gc_2",
    #"starch_to_glc_mal": "MALTODEG_RXN_c_gc_2",
    "mal_deg": "MALIC_NAD_RXN_m_gc_2",
    "mal_v_c": "MAL_vc_gc_2",
    "ps_to_suc": "SUCROSE_PHOSPHATE_SYNTHASE_RXN_c_gc_2",
    #"ps_to_glc": "GLUC1PURIDYLTRANS_RXN_c_gc_2",
    #"cit_to_mal": "FUMHYDR_RXN_m_gc_2",
    "cit_to_mal": "SUCCCOASYN_RXN_m_gc_2",
    "suc_c_v": "SUCROSE_PROTON_cv_gc_2",
    "suc_v_deg": "RXN_1461_v_gc_2",
    "suc_c_deg": "RXN_1461_c_gc_2",
    "suc_to_fru": "SUCROSE_SYNTHASE_RXN_c_gc_2",
    "cit_to_acon": "ACONITATEDEHYDR_RXN_c_gc_2",
    "cit_synth": "CITSYN_RXN_m_gc_2",
    
}

flux_dict = {
    "blue": {
        "unconstrained": {
            "wt": dict(reaction_dict),
            "starchko": dict(reaction_dict)
        },
        "constrained": {
            "wt": dict(reaction_dict),
            "starchko": dict(reaction_dict)
        }
    },
    "white": {
        "constrained": {
            "wt": dict(reaction_dict),
            "starchko": dict(reaction_dict)
        }
    },
    "nops": {
        "constrained": {
            "wt": dict(reaction_dict),
            "starchko": dict(reaction_dict)
        }
    }
}

conditions_list_of_lists = [
    ["blue", "white", "nops"],
    ["unconstrained", "constrained"],
    ["wt", "starchko"],
]
conditions_iterations = list(itertools.product(*conditions_list_of_lists))

solutions_df

Light	blue				white				nops
ATPase	unconstrained		constrained		unconstrained		constrained		unconstrained		constrained
Starch	wt	starchko	wt	starchko	wt	starchko	wt	starchko	wt	starchko	wt	starchko
Reaction
EX_X_pi_t_me	0.000000	0.000000e+00	0.000000e+00	0.000000	0.000000e+00	0.000000	0.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000
EX_X_pi_t_gc	0.000000	0.000000e+00	0.000000e+00	0.000000	0.000000e+00	0.000000	0.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000
EX_X_CO2_t_me	4.128357	4.128357e+00	4.128356e+00	4.128355	-5.388402e+01	-53.884016	-53.884016	-5.388402e+01	-53.884046	-53.884046	-53.884048	-53.884048
EX_X_CO2_t_gc	0.003663	4.036220e-03	2.176874e-03	0.001631	1.156296e-03	0.001168	-0.001087	-1.371060e-03	0.002922	0.002922	0.001837	0.002173
EX_X_Mg_t_me	0.000000	0.000000e+00	0.000000e+00	0.000000	0.000000e+00	0.000000	0.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000
...	...	...	...	...	...	...	...	...	...	...	...	...
MAL_total_pseudolinker	0.004445	4.525811e-03	2.358282e-03	0.001766	4.455754e-03	0.004487	0.000000	-6.494415e-10	0.002314	0.002314	0.000918	0.002355
FRU_total_pseudolinker	0.000000	0.000000e+00	1.737018e-03	0.006863	0.000000e+00	0.000000	0.001898	6.805909e-03	0.000347	0.000347	0.004141	0.006863
K_total_pseudolinker	0.013250	1.303211e-02	9.057052e-03	0.005820	1.292071e-02	0.012970	0.005393	4.495170e-03	0.009849	0.009849	0.006187	0.005980
Cl_total_pseudolinker	0.011572	1.303197e-02	7.424397e-03	0.004597	1.292071e-02	0.012970	0.005393	4.495170e-03	0.008247	0.008247	0.004350	0.004350
CIT_total_pseudolinker	0.000000	4.690523e-08	-4.632768e-10	0.000000	-2.844564e-18	0.000000	0.000000	0.000000e+00	0.000000	0.000000	0.000000	0.000000

1774 rows × 12 columns

for light, light_dict in flux_dict.items():
    for atpase, atpase_dict in light_dict.items():
        for starch, starch_dict in atpase_dict.items():
            for shorthand, reaction_name in starch_dict.items():
                flux = solutions_df.loc[reaction_name[:-2],(light, atpase, starch)]
                if abs(flux) < 1e-8:
                    flux = 0
                flux_dict[light][atpase][starch][shorthand] = flux

solutions_df.loc["GLUC1PURIDYLTRANS_RXN_c_gc"]

Light  ATPase         Starch  
blue   unconstrained  wt          1.190563e-04
                      starchko    1.960967e-04
       constrained    wt         -4.632788e-10
                      starchko    2.462117e-12
white  unconstrained  wt          4.278484e-04
                      starchko    4.304933e-04
       constrained    wt         -9.054699e-04
                      starchko    1.142550e-04
nops   unconstrained  wt         -1.310427e-04
                      starchko   -1.310166e-04
       constrained    wt          0.000000e+00
                      starchko    0.000000e+00
Name: GLUC1PURIDYLTRANS_RXN_c_gc, dtype: float64

flux_dict

{'blue': {'unconstrained': {'wt': {'starch_to_glc': 0,
    'starch_to_mal': 0,
    'mal_deg': 0,
    'mal_v_c': 0.0084862683952195,
    'ps_to_suc': 0.0001190562980565,
    'cit_to_mal': 0.0013558713124941,
    'suc_c_v': 0,
    'suc_v_deg': 0,
    'suc_c_deg': 0,
    'suc_to_fru': 0,
    'cit_to_acon': 0.0013558713124941,
    'cit_synth': 0.000237558821619},
   'starchko': {'starch_to_glc': 0,
    'starch_to_mal': 0,
    'mal_deg': 0.0006650467732122,
    'mal_v_c': 0.010648966832743,
    'ps_to_suc': 0.0001960966512218,
    'cit_to_mal': 0.0008129115147968,
    'suc_c_v': 0,
    'suc_v_deg': 0,
    'suc_c_deg': 0,
    'suc_to_fru': 0,
    'cit_to_acon': 0,
    'cit_synth': 0.0008130053252513}},
  'constrained': {'wt': {'starch_to_glc': 0.0051260186331311,
    'starch_to_mal': 0,
    'mal_deg': 0,
    'mal_v_c': 0.0036281277404994,
    'ps_to_suc': 0,
    'cit_to_mal': 0.0010884364690379,
    'suc_c_v': 0,
    'suc_v_deg': 0.0034740356910968,
    'suc_c_deg': 0,
    'suc_to_fru': 0,
    'cit_to_acon': 0.0010884373955896,
    'cit_synth': 0},
   'starchko': {'starch_to_glc': 0,
    'starch_to_mal': 0,
    'mal_deg': 0,
    'mal_v_c': 0.0027176388646845,
    'ps_to_suc': 0,
    'cit_to_mal': 0.0008152916495569,
    'suc_c_v': 0,
    'suc_v_deg': 0.013726072962268,
    'suc_c_deg': 0,
    'suc_to_fru': 0,
    'cit_to_acon': 0.0008152916544811,
    'cit_synth': 0}}},
 'white': {'constrained': {'wt': {'starch_to_glc': 0.0054174157446452,
    'starch_to_mal': 0,
    'mal_deg': 0,
    'mal_v_c': 0,
    'ps_to_suc': 0.0010865638814297,
    'cit_to_mal': 0,
    'suc_c_v': 0,
    'suc_v_deg': 0.0018046775864697,
    'suc_c_deg': 0,
    'suc_to_fru': 0.0019920337826212,
    'cit_to_acon': 0,
    'cit_synth': 0},
   'starchko': {'starch_to_glc': 0,
    'starch_to_mal': 0,
    'mal_deg': 0,
    'mal_v_c': 0,
    'ps_to_suc': 0.0001142549771026,
    'cit_to_mal': 0,
    'suc_c_v': 0,
    'suc_v_deg': 0.0136118179802412,
    'suc_c_deg': 0,
    'suc_to_fru': 0,
    'cit_to_acon': 0,
    'cit_synth': 0}}},
 'nops': {'constrained': {'wt': {'starch_to_glc': 0.0025691232468596,
    'starch_to_mal': 0.0003061234489482,
    'mal_deg': 0,
    'mal_v_c': 0,
    'ps_to_suc': 0,
    'cit_to_mal': 0.0012244940670225,
    'suc_c_v': 0.0029878403954816,
    'suc_v_deg': 0.0029878403954816,
    'suc_c_deg': 0.0052938626191942,
    'suc_to_fru': 0,
    'cit_to_acon': 0.0012244937957926,
    'cit_synth': 0},
   'starchko': {'starch_to_glc': 0,
    'starch_to_mal': 0,
    'mal_deg': 0,
    'mal_v_c': 0.003622478194246,
    'ps_to_suc': 0,
    'cit_to_mal': 0.0010867432486121,
    'suc_c_v': 0.0061093909571672,
    'suc_v_deg': 0.0096469801196154,
    'suc_c_deg': 0.0040790928377284,
    'suc_to_fru': 0,
    'cit_to_acon': 0.0010867432486121,
    'cit_synth': 0}}}}

from sigfig import round

def NestedDictValues(d):
    for v in d.values():
        if isinstance(v, dict):
            yield from NestedDictValues(v)
        else:
            yield v

min_flux = np.array(list(NestedDictValues(flux_dict))).min()
min_flux

0.0

max_flux = np.array(list(NestedDictValues(flux_dict))).max()
max_flux

0.013726072962268

max_flux/min_flux

/tmp/ipykernel_4834/4031404064.py:1: RuntimeWarning: divide by zero encountered in scalar divide
  max_flux/min_flux

inf

So we have a 50-fold difference between the smallest and largest flux. 5mm in the inkscape figures that I’ve created is probably the max I can do, so 0.1 is the minimum.

Inkscape will only do widths to nearest 0.01mm I think, but allows you to input 3 numbers after the dp.

def convert_fluxes_to_widths_linear(flux):
    max_width_in_inkscape = 5
    conversion = max_width_in_inkscape/max_flux

    new_flux = conversion*flux
    new_flux_rounded = round(new_flux, 3)

    return new_flux_rounded

import matplotlib.pyplot as plt

x = np.arange(0.0001, 0.014, 0.0001)
y = [convert_fluxes_to_widths_linear(flux) for flux in x]

plt.plot(x, y)

import copy

def convert_flux_dict(flux_dict):
    converted_dict = copy.deepcopy(flux_dict)
    for light_value, light_dict in flux_dict.items():
        for atpase_constraint, atpase_dict in light_dict.items():
            for phenotype, phenotype_dict in atpase_dict.items():
                for reaction, reaction_value in phenotype_dict.items():
                    converted_dict[light_value][atpase_constraint][phenotype][reaction] = convert_fluxes_to_widths_linear(
                        reaction_value)

    return converted_dict

converted_dict = convert_flux_dict(flux_dict)

/home/maurice/miniconda3/envs/mmon-gcm/lib/python3.9/site-packages/sigfig/sigfig.py:586: UserWarning: warning: 3 significant figures requested from number with only 1 significant figures
  warn("warning: %d significant figures requested from number with only %d significant figures" % (given['sigfigs'], len(num.map)))
/home/maurice/miniconda3/envs/mmon-gcm/lib/python3.9/site-packages/sigfig/sigfig.py:586: UserWarning: warning: 3 significant figures requested from number with only 2 significant figures
  warn("warning: %d significant figures requested from number with only %d significant figures" % (given['sigfigs'], len(num.map)))

for light, light_dict in converted_dict.items():
    for atpase, atpase_dict in light_dict.items():
        for starch, starch_dict in atpase_dict.items():
            print(f"{light}_{atpase}_{starch}:")
            for shorthand, mm in starch_dict.items():
                if mm > 0:
                    #print(f"{shorthand} should be {mm}mm wide")
                    print(f"{shorthand} should be {mm*175}%")
            print("\n")

blue_unconstrained_wt:
mal_v_c should be 540.75%
ps_to_suc should be 7.595%
cit_to_mal should be 86.45%
cit_to_acon should be 86.45%
cit_synth should be 15.1375%


blue_unconstrained_starchko:
mal_deg should be 42.35%
mal_v_c should be 679.0%
ps_to_suc should be 12.495000000000001%
cit_to_mal should be 51.8%
cit_synth should be 51.8%


blue_constrained_wt:
starch_to_glc should be 327.25%
mal_v_c should be 231.0%
cit_to_mal should be 69.3%
suc_v_deg should be 222.25%
cit_to_acon should be 69.3%


blue_constrained_starchko:
mal_v_c should be 173.25%
cit_to_mal should be 51.974999999999994%
suc_v_deg should be 875.0%
cit_to_acon should be 51.974999999999994%


white_constrained_wt:
starch_to_glc should be 344.75%
ps_to_suc should be 69.3%
suc_v_deg should be 114.97500000000001%
suc_to_fru should be 127.05%


white_constrained_starchko:
ps_to_suc should be 7.279999999999999%
suc_v_deg should be 868.0%


nops_constrained_wt:
starch_to_glc should be 163.8%
starch_to_mal should be 19.6%
cit_to_mal should be 78.05%
suc_c_v should be 190.75%
suc_v_deg should be 190.75%
suc_c_deg should be 337.75%
cit_to_acon should be 78.05%


nops_constrained_starchko:
mal_v_c should be 231.0%
cit_to_mal should be 69.3%
suc_c_v should be 390.25%
suc_v_deg should be 614.25%
suc_c_deg should be 260.75%
cit_to_acon should be 69.3%

For the key:

import cobra
import pandas as pd
from mmon_gcm.supermodel import SuperModel

parameters_df = pd.read_csv("../inputs/arabidopsis_parameters.csv", index_col=0)
four_stage_GC_model = cobra.io.sbml.read_sbml_model("../models/4_stage_GC.xml")  # read model
arabidopsis_supermodel = SuperModel(parameters_df.loc[:, "Value"], fba_model=four_stage_GC_model);

No objective coefficients in model. Unclear what should be optimized

arabidopsis_supermodel.N_gcs

580000000.0

0.01/arabidopsis_supermodel.N_gcs

1.724137931034483e-11

moles_gc_h = 10 * 10 ** -15  # fmolesgc-1h-1

moles_m2_h = moles_gc_h*arabidopsis_supermodel.N_gcs  # molesh-1

mmoles_m2_h = moles_m2_h * 10 ** 3
mmoles_m2_h

0.005800000000000001

f"{convert_fluxes_to_widths_linear(mmoles_m2_h)}mm for a flux of 10 fmoles gc-1 h-1"

'2.11mm for a flux of 10 fmoles gc-1 h-1'