data-processing-pipeline/model.py at main · ambrosia2024/data-processing-pipeline · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

# --- Secondary Models from FSKX ---
def get_umax(temp, b, Tmin):
    return (b * (temp - Tmin)) ** 2

def get_lag(temp, b, Tmin, alpha0):
    log10 = np.log(10)
    umax = get_umax(temp, b, Tmin)
    return -np.log(alpha0) / (umax * log10)

def get_ymax(temp):
    # Linear model for maximum population density from the paper:
    # MPD(xmax, ln CFU/g) = 0.037T + 12.434
    # Converting from ln to log10 by dividing by ln(10)
    return (0.037 * temp + 12.434) / np.log(10)

# --- Primary Growth Model: Baranyi & Roberts (1994) ---
def listeria_growth_model(t, y0, ymax, umax, lag):
    log10 = np.log(10)
    term_exp1 = np.exp(-umax * log10 * t)
    term_exp2 = np.exp(-umax * log10 * (t - lag))
    log_part = np.log(1 - term_exp1 + term_exp2)
    term1 = umax * (t - lag + (log_part / (umax * log10)))
    growth_exponent = umax * log10 * (t - lag + (log_part / (umax * log10)))
    numerator = np.exp(growth_exponent) - 1
    denominator = 10 ** ((ymax - y0))
    term2 = (1 / (log10)) * np.log(1 + (numerator / denominator))
    Nt = y0 + term1 - term2
    return Nt

# --- Embedded Experimental Data as DataFrames ---
# Updated with actual Listeria data from ComBase export
data = {
    5: pd.DataFrame([
        (0.0, 3.38), (24.0, 3.36), (48.0, 3.61), (72.0, 3.66),
        (96.0, 4.21), (120.0, 4.50), (144.0, 4.36), (168.0, 4.66)
    ], columns=["Time", "LogCFU"]),
    10: pd.DataFrame([
        (0.0, 3.43), (24.0, 3.43), (48.0, 3.62), (54.0, 3.85),
        (60.0, 4.12), (72.0, 4.64), (96.0, 4.81), (120.0, 4.81),
        (144.0, 4.81), (168.0, 4.80)
    ], columns=["Time", "LogCFU"]),
    15: pd.DataFrame([
        (0.0, 3.37), (4.0, 3.32), (5.0, 3.36), (6.0, 3.33),
        (7.0, 3.43), (8.0, 3.59), (9.0, 3.58), (10.0, 3.57),
        (11.0, 3.66), (12.0, 3.62), (13.0, 3.58), (14.0, 3.77),
        (15.0, 3.94), (16.0, 3.87), (17.0, 4.08), (18.0, 3.91),
        (19.0, 4.39), (20.0, 4.20), (22.0, 4.34), (23.0, 4.42),
        (24.0, 4.78), (25.0, 4.78), (36.0, 4.88)
    ], columns=["Time", "LogCFU"]),
    20: pd.DataFrame([
        (0.0, 3.48), (1.0, 3.17), (2.0, 3.39), (3.0, 3.38),
        (4.0, 3.43), (5.0, 3.64), (6.0, 3.69), (7.0, 3.66),
        (8.0, 4.02), (9.0, 4.03), (10.0, 4.08), (11.0, 4.29),
        (12.0, 4.33), (13.0, 4.74), (14.0, 4.73), (15.0, 4.89),
        (16.0, 5.13), (17.0, 5.12), (18.0, 5.13), (19.0, 5.22),
        (23.0, 5.18), (28.0, 5.20), (36.0, 5.20)
    ], columns=["Time", "LogCFU"]),
    25: pd.DataFrame([
        (0.0, 3.56), (1.0, 3.20), (2.0, 3.30), (3.0, 3.55),
        (4.0, 3.42), (5.0, 3.39), (6.0, 3.85), (7.0, 4.10),
        (8.0, 4.32), (9.0, 4.38), (10.0, 4.44), (11.0, 4.82),
        (12.0, 4.82), (13.0, 5.14), (14.0, 5.09), (15.0, 5.16),
        (28.0, 5.18), (36.0, 5.20)
    ], columns=["Time", "LogCFU"])
}

# --- Simulation Function ---
def simulate_growth(temp_celsius, y0, tmax,  b, Tmin, alpha0):
    t = np.linspace(0, tmax, 100)
    umax = get_umax(temp_celsius, b, Tmin)
    lag = get_lag(temp_celsius, b, Tmin, alpha0)
    ymax = get_ymax(temp_celsius)
    model_values = listeria_growth_model(t, y0, ymax, umax, lag)
    return t, model_values


# Run simulation with default parameters
# t, model_values = simulate_growth(temp_celsius, y0, tmax, b, Tmin, alpha0)