4mCPred-GSIMP: Predicting DNA N4-methylcytosine sites in the mouse genome with multi-Scale adaptive features extraction and fusion

Jianhua Jia; Yu Deng; Mengyue Yi; Yuhui Zhu; Jianhua Jia; Yu Deng; Mengyue Yi; Yuhui Zhu

doi:10.3934/mbe.2024012

Mathematical Biosciences and Engineering

2024, Volume 21, Issue 1: 253-271. doi: 10.3934/mbe.2024012

Previous Article Next Article

Research article Special Issues

4mCPred-GSIMP: Predicting DNA N4-methylcytosine sites in the mouse genome with multi-Scale adaptive features extraction and fusion

School of Information Engineering, Jingdezhen Ceramic University, Jingdezhen 333403, China

Academic Editor: Qianqian Song

Received: 12 October 2023 Revised: 30 November 2023 Accepted: 04 December 2023 Published: 12 December 2023

The epigenetic modification of DNA N4-methylcytosine (4mC) is vital for controlling DNA replication and expression. It is crucial to pinpoint 4mC's location to comprehend its role in physiological and pathological processes. However, accurate 4mC detection is difficult to achieve due to technical constraints. In this paper, we propose a deep learning-based approach 4mCPred-GSIMP for predicting 4mC sites in the mouse genome. The approach encodes DNA sequences using four feature encoding methods and combines multi-scale convolution and improved selective kernel convolution to adaptively extract and fuse features from different scales, thereby improving feature representation and optimization effect. In addition, we also use convolutional residual connections, global response normalization and pointwise convolution techniques to optimize the model. On the independent test dataset, 4mCPred-GSIMP shows high sensitivity, specificity, accuracy, Matthews correlation coefficient and area under the curve, which are 0.7812, 0.9312, 0.8562, 0.7207 and 0.9233, respectively. Various experiments demonstrate that 4mCPred-GSIMP outperforms existing prediction tools.

Keywords:

Citation: Jianhua Jia, Yu Deng, Mengyue Yi, Yuhui Zhu. 4mCPred-GSIMP: Predicting DNA N4-methylcytosine sites in the mouse genome with multi-Scale adaptive features extraction and fusion[J]. Mathematical Biosciences and Engineering, 2024, 21(1): 253-271. doi: 10.3934/mbe.2024012

Related Papers:

[1]	Shaimaa A. M. Abdelmohsen, D. Sh. Mohamed, Haifa A. Alyousef, M. R. Gorji, Amr M. S. Mahdy . Mathematical modeling for solving fractional model cancer bosom malignant growth. AIMS Biophysics, 2023, 10(3): 263-280. doi: 10.3934/biophy.2023018
[2]	Mati ur Rahman, Mehmet Yavuz, Muhammad Arfan, Adnan Sami . Theoretical and numerical investigation of a modified ABC fractional operator for the spread of polio under the effect of vaccination. AIMS Biophysics, 2024, 11(1): 97-120. doi: 10.3934/biophy.2024007
[3]	Yasir Nadeem Anjam, Mehmet Yavuz, Mati ur Rahman, Amna Batool . Analysis of a fractional pollution model in a system of three interconnecting lakes. AIMS Biophysics, 2023, 10(2): 220-240. doi: 10.3934/biophy.2023014
[4]	Marco Menale, Bruno Carbonaro . The mathematical analysis towards the dependence on the initial data for a discrete thermostatted kinetic framework for biological systems composed of interacting entities. AIMS Biophysics, 2020, 7(3): 204-218. doi: 10.3934/biophy.2020016
[5]	Mehmet Yavuz, Kübra Akyüz, Naime Büşra Bayraktar, Feyza Nur Özdemir . Hepatitis-B disease modelling of fractional order and parameter calibration using real data from the USA. AIMS Biophysics, 2024, 11(3): 378-402. doi: 10.3934/biophy.2024021
[6]	Larisa A. Krasnobaeva, Ludmila V. Yakushevich . On the dimensionless model of the transcription bubble dynamics. AIMS Biophysics, 2023, 10(2): 205-219. doi: 10.3934/biophy.2023013
[7]	Carlo Bianca . Differential equations frameworks and models for the physics of biological systems. AIMS Biophysics, 2024, 11(2): 234-238. doi: 10.3934/biophy.2024013
[8]	Mohammed Alabedalhadi, Mohammed Shqair, Ibrahim Saleh . Analysis and analytical simulation for a biophysical fractional diffusive cancer model with virotherapy using the Caputo operator. AIMS Biophysics, 2023, 10(4): 503-522. doi: 10.3934/biophy.2023028
[9]	Bertrand R. Caré, Pierre-Emmanuel Emeriau, Ruggero Cortini, Jean-Marc Victor . Chromatin epigenomic domain folding: size matters. AIMS Biophysics, 2015, 2(4): 517-530. doi: 10.3934/biophy.2015.4.517
[10]	David Gosselin, Maxime Huet, Myriam Cubizolles, David Rabaud, Naceur Belgacem, Didier Chaussy, Jean Berthier . Viscoelastic capillary flow: the case of whole blood. AIMS Biophysics, 2016, 3(3): 340-357. doi: 10.3934/biophy.2016.3.340

Abstract

1. Introduction

Numerous mathematical models have been established to predict and study the biological system. In the past four decades, there have been far-reaching research on improving cell mass production in chemical reactors [1]. The chemostat model is used to understand the mechanism of cell mass growth in a chemostat. A chemostat is an apparatus for continuous culture that contains bacterial populations. It can be used to investigate the cell mass production under controlled conditions. This reactor provides a dynamic system for population studies and is suitable to be used in a laboratory. A substrate is continuously added into the reactor containing the cell mass, which grows by consuming the substrate that enters through the inflow chamber. Meanwhile, the mixture of cell mass and substrate is continuously harvested from the reactor through the outflow chamber. The dynamics in the chemostat can be investigated by using the chemostat model [2].

Ordinary differential equations (ODEs) are commonly used for modelling biological systems. However, most biological systems behaviour has memory effects, and ODEs usually neglect such effects. The fractional-order differential equations (FDEs) are taken into account when describing the behaviour of the systems' equations. A FDEs is a generalisation of the ODEs to random nonlinear order [3]. This equation is more effective because of its good memory, among other advantages [4]–[8]. The errors occuring from the disregarded parameters when modelling of phenomena in real-life also can be reduced. FDEs are also used to efficiently replicate the real nature of various systems in the field of engineering and sciences [9]. In the past few decades, FDEs have been used in biological systems for various studies [5], [6], [10]–[17].

Since great strides in the study of FDEs have been developed, the dynamics in the chemostat can be investigated using the mathematical model of the chemostat in the form of FDEs. Moreover, there have been few studies on the expansion of the chemostat model with fractional-order theory. Thus, we deepen and complete the analysis on the integer-order chemostat model with fractional-order theory and discuss the stability of the equilibrium points of the fractional-order chemostat model. Next, the bifurcation analysis for the fractional-order chemostat model is conducted to identify the bifurcation point that can change the stability of the system. The analysis identifies the values of the fractional-order and the system parameters to ensure the operation of the chemostat is well-controlled.

2. Materials and methods

2.1. Definition of Caputo derivative

Recently, there are many approaches to define fractional operators such as by Caputo, Riemann-Liouville, Hadamard, and Grunwald-Letnikov [3]. However, Caputo is often used due to its convenience in various applications [18]. Caputo is also useful to encounter an obstacle where the initial condition is done in the differential of integer-order [19]. In this paper, we applied Caputo derivative to define the system of fractional-order. The Caputo derivative for the left-hand side is defined as

$D_{t}^{α} f (t) = \frac{1}{Γ (n - α)} \int_{0}^{t} \frac{f^{(n)} τ}{{(t - τ)}^{α - n + 1}} d τ,$ (2.1)

where Г denotes the function of gamma, n is an integer, where $n - 1 < α < n$ [18].

2.2. Adams-type predictor corrector method

The Adams-type predictor-corrector method is one of the technique that have been proposed for fractional-order differential equations [19]–[21]. The Adams-type predictor-corrector method is a analysis of numerical algorithm that involves two basics steps: predictor and corrector. The predictor formula can be described as

$\begin{array}{l} y_{h}^{P} (t_{n + 1}) = \sum_{k = 0}^{α - 1} \frac{t_{n + 1}^{k}}{k!} y_{0}^{(k)} + \frac{1}{Γ (α)} \sum_{j = 0}^{n} b_{j, n + 1} f (t_{j}, y_{h} (t_{j})), \end{array}$ (2.2)

meanwhile the corrector formula can be determined by

$y_{h} (t_{n + 1}) = \sum_{k = 0}^{α - 1} \frac{t_{n + 1}^{k}}{k!} + \frac{h^{α}}{Γ (α + 2)} f (t_{n + 1}, y_{h}^{P} (t_{n + 1})) + \frac{h^{α}}{Γ (α + 2)} \sum_{j = 0}^{n} f (t_{j}, y_{h} (t_{j})) .$ (2.3)

The predictor-corrector method is also called as the PECE (Predict, Evaluate, Correct, Evaluate) method [22]. The procedure of the predictor-corrector method can be explained as follows

(i) Calculate the predictor step, $y_{h}^{P} (t_{n + 1})$ in Eq. (2.2).

(ii) Evaluate $f (t_{n + 1}, y_{h}^{P} (t_{n + 1}))$ .

(iii) Calculate the corrector step, $y_{h} (t_{n + 1})$ in Eq. (2.3).

(iv) Evaluate $f (t_{n + 1}, y_{h} (t_{n + 1}))$ .

The procedure repeatedly predicts and corrects the value until the corrected value becomes a converged number [21]. This method is able to maintain the stability of the properties and has good accuracy. Moreover, this method also has lower computational cost than other methods [23]. The algorithm of the Adams-type predictor-corrector method proposed by [22] is shown in Appendix.

2.3. Stability analysis of fractional-order system

The conditions of stability for integer-order differential equations and fractional-order differential equations are different. Both systems could have the same steady-state points but different stability conditions [24], [25]. The stability condition for fractional-order differential equations can be stated by Theorem 1 and the Routh-Hurwitz stability condition as described by Proposition 1.

Theorem 1 [5], [6], [26]. The commensurate system of fractional-order where $x \in ℝ$ and $0 < α < 1$ is locally asymptotically stable if the eigenvalues of the Jacobian matrix evaluated at the steady-state point is satisfied by

$| a r g (λ) | > \frac{α π}{2} .$ (2.4)

Proposition 1 [5], [6], [26]. Suppose the characteristic polynomial is $P (λ) = λ^{2} + b λ + c$ of the Jacobian matrix which evaluated at the steady-state. The eigenvalues of the Jacobian matrix will satisfy Eq. (2.4) in Theorem 1 if

$b > 0, c > 0,$ (2.5)

$b < 0, 4 c > b^{2}, | t a n^{- 1} (\frac{\sqrt{4 c - b^{2}}}{b}) | > \frac{α π}{2} .$ (2.6)

The stability theorem on the fractional-order systems and fractional Routh-Hurwitz stability conditions are introduced to analyze the stability of the model. The fractional Routh-Hurwitz stability conditions is specifically introduced for the eigenvalues of the Jacobian matrix that obtained in quadratic form. The proof of this proposition is shown in Appendix.

2.4. Hopf bifurcation analysis of fractional-order system

Bifurcation can be defined as any sudden change that occurs while a parameter value is varied in the differential equation system and it has a significant influence on the solution [27]. An unstable steady-state may becomes stable and vice versa. A slight changes in the parameter value may change the system's stability. Despite the steady-state point and the eigenvalues of the system of fractional-order are similar as the system of integer-order, the discriminant method used for the stability of the steady-state point is different. Accordingly, the Hopf bifurcation condition in the fractional-order system is slightly different as compared with the integer-order system.

2.4.1. Hopf bifurcation analysis of fractional order α

Fractional order α can be selected as the bifurcation parameter in a fractional-order system, but this is not allowed in an integer-order system. The existence of Hopf bifurcation can be stated as in Theorem 2.

Theorem 2 [28]. Assume α* as the critical value of the fractional-order. When bifurcation parameter α passes over critical value α*, which is $α^{*} \in (0,1)$ , Hopf bifurcation occurs at the steady-state point if the following conditions are satisfied

(i) 1. The characteristic equation of chemostat system has a pair of complex conjugate roots, $λ_{1,2} = p \pm i q$ , while the other eigenvalues are negative real roots.

(ii) Critical value $m (α^{*}) = \frac{α^{*} π}{2} - m i n | a r g (λ) | = 0$ .

(iii) $\frac{d m (α)}{d α} |_{α = α^{*}} \neq 0$ (condition of transversality).

Proof. Condition (i) is not easy to obtain due to the selected parameter's value. However, this condition can be managed under some confined conditions. In fact, the washout steady-state solution of the chemostat model has two negative real roots. The remaining two roots depend on the characteristic of the polynomial from the no-washout steady-state solution.

Condition (ii) can be satisfied with the existence of critical value α* and when $a r g (λ)$ is equivalent to $a r c t a n (\frac{q}{p})$ . Thus, the solution of critical value $m (α^{*})$ can be written as

$α^{*} = \frac{α^{*} π}{2} - a r c t a n (\frac{q}{p}) = 0, α^{*} \in (0,1) .$ (2.7)

The integer system required p = 0 for the bifurcation's operating condition. For the fractional-order system, the operating condition of the system will change into $m (α^{*}) = \frac{α^{*} π}{2} - m i n | a r g (x) | = 0$ . For condition (iii), the condition of $m (α)$ changes when bifurcation parameter $α$ passes over critical value α*. For example, the steady-state point is asymptotically stable for $0 < α < α^{*}$ and unstable when $α < α^{*} < 1$ . Thus, Hopf bifurcation exists at $α = α^{*}$ .

2.4.2. Hopf bifurcation analysis of parameters value

In studying the dynamic process of chemostat, the parameters such as $Q, S_{0}, µ, k, γ$ and β are usually used as the bifurcation parameter since these parameters have significant effects on the dynamic process of the system of fractional-order and integer-order. Fractional-order α is considered fixed and the initial substrate concentration S₀ is studied as the control parameter. The existence of the Hopf bifurcation can be stated as in Theorem 3.

Theorem 3 [28]. Assume $S_{0}^{*}$ as the critical value of the fractional-order. When bifurcation parameter S₀ passes over critical value $S_{0}^{*}$ , Hopf bifurcation occurs at the steady-state point if the following conditions are satisfied

(i) The characteristic equation of chemostat system has a pair of complex conjugate roots, $λ_{1,2} = p (S_{0}) \pm i q (S_{0})$ , while the other eigenvalues are negative real roots.

(ii) Critical value $m (S_{0}^{*}) = \frac{α π}{2} - m i n | a r g (λ (S_{0}^{*})) | = 0$ .

(iii) $\frac{d m (S_{0}^{*})}{d (S_{0}^{*})} |_{S_{0} = S_{0}^{*}} \neq 0$ (condition of transversality).

Proof. This theorem can be proved in the same way as Theorem 2. Therefore, condition (i) can be guaranteed. Condition (ii) can be satisfied with the existence of critical value $S_{0}^{*}$ and when $a r g [λ (S_{0}^{*})]$ is equivalent to $a r c t a n [\frac{q (S_{0}^{*})}{p (S_{0}^{*})}]$ . Thus, the solution of critical value $m (S_{0}^{*})$ can be written as

$S_{0}^{*} = \frac{α π}{2} - a r c t a n (\frac{q (S_{0}^{*})}{p (S_{0}^{*})}) = 0.$ (2.8)

For condition (iii), the condition of $m (S_{0}^{*})$ changes when bifurcation parameter S₀ passes over critical value $S_{0}^{*}$ . For example, the steady-state point is asymptotically stable when $0 < S_{0} < S_{0}^{*}$ and unstable when $S_{0} < S_{0}^{*} < 1$ . Thus, Hopf bifurcation exists at $S_{0} = S_{0}^{*}$ [28].

Firstly, determine the steady-states, Jacobian matrix and eigenvalues of the fractional-order chemostat model. The stability properties of the fractional-order chemostat model were estimated by using the stability and bifurcation analyses with FDEs by referring to Theorem 1 and Proposition 1. Then, determine the bifurcation point of fractional-order by referring to Theorem 2 and determine the bifurcation point of parameter values by referring to Theorem 3. Next, plot the phase portrait of fractional-order chemostat model by using Adam-types predictor-corrector method to study the dynamic behaviour of the system. Figure 1 depicts the flowchart of this research. This flowchart can be applied to all problems with suitable parameter values.

Figure 1. Mathematical analysis of fractional-order chemostat model.

DownLoad: Full-Size Img PowerPoint

3. Results and Discussion

3.1. Fractional-order chemostat model

An integer-order chemostat model that considered a variable yield coefficient and the Monod growth model from [1] is studied in this section. The chemostat system can be written as

$\begin{array}{l} \frac{d S}{d t} & = Q (S_{0} - S) - \frac{µ S X}{(k + S) (γ + β S)}, \\ \frac{d X}{d t} & = Q (- X) + \frac{µ S X}{k + S}, \end{array}$ (3.1)

with the initial value of $X_{0} = 0$ , where the sterile feed case was assumed. The integer-order chemostat system of Eq. (3.1) is extended to the fractional-order differential equation

$\begin{array}{l} \frac{d^{α} S}{d t^{α}} & = Q (S_{0} - S) - \frac{µ S X}{(k + S) (γ + β S)}, \\ \frac{d^{α} X}{d t^{α}} & = Q (- X) + \frac{µ S X}{k + S} . \end{array}$ (3.2)

Let Eq. (3.2) equal to zero in order to find the steady-state solutions

$Q (S_{0} - S) - \frac{µ S X}{(k + S) (γ + β S)} = 0,$ (3.3)

$Q (- X) + \frac{µ S X}{k + S} = 0.$ (3.4)

By solving Eq. (3.4), the following solutions are obtained

$S^{*} = \frac{k Q}{µ - Q},$ (3.5)

$X^{*} = 0.$ (3.6)

From Eq. (3.3), if $X^{*} = 0$ , then $S^{*} = S_{0}$ . If $S^{*} = \frac{k Q}{µ - Q}$ , then

$X^{*} = \frac{(k Q + Q S_{0} - S_{0} µ) (- Q γ + k Q β + γ µ)}{{(µ - Q)}^{2}} .$ (3.7)

Hence, the solutions of steady-state for the chemostat model are

(i) Washout:

$(S_{0}^{*}, X_{0}^{*}) = (S_{0},0) .$ (3.8)

(ii) No Washout:

$(S_{1}^{*}, X_{1}^{*}) = (ρ, (S_{0} - ρ) (γ + β ρ)),$ (3.9)

where

$\begin{array}{l} ρ = \frac{k Q}{µ - Q} . \end{array}$

The steady-state solutions are physically meaningful if their components are positive. Therefore, $S_{0} > 0$ for the washout steady-state solution exists by biological meaning. The no-washout steady-state solution will only exist when $0 < ρ < S_{0}$ . The Jacobian matrix of no washout steady-state as in Eq. (3.10) can be used to investigate the stability properties of the fractional-order chemostat model.

$J = [\begin{matrix} - Q + \frac{X (- k γ + S^{2} β) µ}{{(k + S)}^{2} {(γ + S β)}^{2}} & \frac{S µ}{(k + S) (γ + S β)} \\ \frac{k X µ}{{(k + S)}^{2}} & - Q + \frac{S µ}{k + S} \end{matrix}] .$ (3.10)

3.2. Stability of washout and no-washout steady-state solution

The solution of steady-state in Equation (3.8) represents the washout situation, where the cell mass is wholly removed from the reactor and where the substrate concentration is at the same stock as in the beginning. This state must always be unstable in order to ensure that the cell mass is able to grow in the chemostat. This is because the cell mass will be continuously removed from the chemostat if the washout steady-state is stable. The Jacobian matrix for the washout steady-state solution can be written as

$J = [\begin{matrix} - Q & - \frac{S_{0} µ}{(k + S_{0}) (γ + S_{0} β)} \\ 0 & - Q + \frac{S_{0} µ}{k + S_{0}} \end{matrix}] .$ (3.11)

The eigenvalues of this matrix are

$λ_{1} = - Q,$ (3.12)

$λ_{2} = \frac{- k Q - Q S_{0} + S_{0} µ}{k + S_{0}} .$ (3.13)

The eigenvalues in Eq. (3.12) and Eq. (3.13) are real. The washout steady-state solution is stable if Q>0 and $ρ > S_{0}$ where $ρ = \frac{k Q}{µ - Q}$ . The steady-state solution in Eq. (3.9) represents the no-washout situation. No-washout situation is where the cell mass is not removed and stay growth in the chemostat.This state is important. The steady-state solution is substituted into the Jacobian matrix in Eq. (3.10) and can be written as

$J = [\begin{matrix} - Q + \frac{µ (S_{0} - ρ) (β ρ^{2} - k γ)}{{(k + ρ)}^{2} (β ρ - γ)} & - \frac{µ ρ}{(k + ρ) (β ρ + γ)} \\ \frac{k µ (S_{0} - ρ) (β ρ - γ)}{{(k + ρ)}^{2}} & - Q + \frac{µ ρ}{k + ρ} \end{matrix}] .$ (3.14)

The eigenvalues of the Jacobian matrix in terms of the characteristic polynomial are

$P (λ) = λ^{2} + b λ + c,$ (3.15)

where

$b = 2 Q + \frac{µ (S_{0} - 2 ρ)}{(k + ρ)} + \frac{µ ρ (ρ - S_{0})}{{(k + ρ)}^{2}} + \frac{β µ ρ (ρ - S_{0})}{(k + ρ) (γ + β ρ)},$ (3.16)

and

$\begin{array}{l} c = & Q^{2} + \frac{Q µ (S_{0} - 2 ρ)}{(k + ρ)} + \frac{(µ ρ - S_{0} µ) (µ ρ - Q ρ)}{{(k + ρ)}^{2}} + \frac{µ^{2} ρ^{2} (S_{0} - ρ)}{{(k + ρ)}^{3}} \\ - \frac{Q β µ ρ (S_{0} - ρ)}{(k + ρ) (γ + β ρ)} + \frac{(S_{0} µ - µ ρ) (µ ρ γ + 2 µ β ρ^{2})}{{(k + ρ)}^{2} (γ + β ρ)} \\ + \frac{(µ ρ - S_{0} µ) (µ ρ^{2} γ + µ β ρ^{3})}{{(k + ρ)}^{3} (γ + β ρ)} . \end{array}$ (3.17)

The eigenvalues of the no-washout steady-state solution were evaluated with Routh-Hurwitz condition in Proposition 1. Based on the eigenvalues in Eq. (3.15) and by referring to the study by [5], the eigenvalues' condition can be simplified as the following two cases

(i) If b>0 or equivalent to $\frac{γ}{β} > P_{1}$ , the no-washout steady-state solution of the system in Eq. (3.2) is asymptotically stable. P₁ can be written as

$P_{1} = - \frac{µ ρ (ρ - S_{0})}{2 Q (k + ρ)} - \frac{ρ (ρ - S_{0})}{(S_{0} - ρ)} - \frac{(k + ρ)}{µ ρ (ρ - S_{0})} - ρ,$ (3.18)

(ii) If b<0 or equivalent to $\frac{γ}{β} < P_{1}$ and $t a n^{- 1} (\frac{\sqrt{4 c - b^{2}}}{b}) > \frac{α π}{2}$ , the no-washout steady-state solution of the system in Eq. (3.2) is asymptotically stable. The condition of $t a n^{- 1} (\frac{\sqrt{4 c - b^{2}}}{b}) > \frac{α π}{2}$ is also equivalent to $4 c o s^{2} (\frac{α π}{2}) c > b^{2}$ , which can be simplified as $\frac{γ}{β} > P_{2}$ . Then, this case can be concluded and written as $P_{2} < \frac{γ}{β} < P_{1}$ where

$P_{2} = \frac{µ ρ (ρ - S_{0})}{2 c o s (\frac{α π}{2}) \sqrt{c} (k + ρ)} - \frac{µ ρ (ρ - S_{0})}{2 Q (k + ρ)} - \frac{ρ (ρ - S_{0})}{(S_{0} - ρ)} - k - 2 ρ .$ (3.19)

Then, if $P_{2} < \frac{γ}{β} < P_{1}$ , the no-washout steady-state solution of the system in Eq. (3.2) is asymptotically stable.

The parameter values of the fractional-order chemostat model are provided in Table 1. The initial substrate concentration, S₀ and ρ were assumed as non-negative values to ensure that the steady-state solutions were physically meaningful. The stability diagram of the steady-state solutions is plotted in Figure 2.

Table 1. Parameter values.

Parameters	Description	Values	Units
k	Saturation constant	1.75	gl⁻¹
Q	Dilution rate	0.02	l²gr⁻¹
µ	Maximum growth rate	0.3	h⁻¹
γ	Constant in yield coefficient	0.01	–
β	Constant in yield coefficient	5.25	lg⁻¹
S₀	Input concentration of substrate	1	gl⁻¹

| Show Table

DownLoad: CSV

Figure 2. Stability diagram of the steady-state solutions when $α = 1$ .

DownLoad: Full-Size Img PowerPoint

The washout steady-state solution is stable if Q > 0 and $ρ > S_{0}$ . From Figure 2, it shows that the unstable solution of washout steady-state, as the eigenvalues did not fulfil the condition of $ρ > S_{0}$ . By choosing the appropriate parameter values, the unstable washout steady-state solution could ensure that the washout condition does not occur in the chemostat. Meanwhile, the solution of no-washout steady-state is stable.

3.3. Hopf bifurcation analysis of the order of fractional-order system

The steady-state solutions of the fractional-order chemostat model for the parameter values given in Table 1 are

(i) Washout:

$(S_{0}^{*}, X_{0}^{*}) = (1,0),$ (3.20)

(ii) No Washout:

$(S_{1}^{*}, X_{1}^{*}) = (\frac{1}{8}, \frac{3731}{6400}) .$ (3.21)

The eigenvalues obtained from the washout steady-state solution are

$λ_{1} = - \frac{1}{50},$ (3.22)

$λ_{2} = - \frac{49}{550},$ (3.23)

and the eigenvalues from the no-washout steady-state solution are

$λ_{1} = \frac{- 2552 + \sqrt{411098126 i}}{399750},$ (3.24)

$λ_{2} = \frac{- 2552 - \sqrt{411098126 i}}{399750},$ (3.25)

Based on Eq. (3.22) to Eq. (3.25), these satisfied the first condition of Hopf bifurcation in Theorem 2. There exists a pair of complex conjugate roots and the other eigenvalues are negative real roots. The transversality condition as the third condition is also satisfied. The eigenvalues of the washout steady-state solution based on the chemostat system is not imaginary, and so there is no existence of Hopf bifurcation in the washout steady-state solution. According to Theorem 2, the critical value of the fractional-order as stated in the second condition can be obtained as

$m (α^{*}) = \frac{α^{*} π}{2} - m i n | a r g (λ) | = 0,$ (3.26)

$α^{*} = \frac{2}{π} m i n | a r g (λ) |,$ (3.27)

where

$a r g (λ) = a r c t a n (\frac{q}{p}),$ (3.28)

$α^{*} = \frac{2}{π} a r c t a n (\frac{q}{p}) = \frac{2}{π} a r c t a n (\frac{\frac{\sqrt{411098126}}{399750}}{\frac{2552}{399750}}) = 0.9202904711 \approx 0.9.$ (3.29)

Value of p and q are obtained from Eq. (3.24) and Eq. (3.25) by assuming parameter value in Table 1. Hence, when $α^{*} = 0.9$ , the chemostat system in Eq. (3.2) shows Hopf bifurcation, at which the system stability would be altered.

Figure 3 is plotted to determine the dynamic behaviour at the Hopf bifurcation point. The phase portrait diagrams of cell mass concentration against substrate concentration are plotted for values of order of the fractional is $α = 0.9$ .

Figure 3. Phase portrait plot of fractional-order chemostat system with $α = 0.9$ .

DownLoad: Full-Size Img PowerPoint

The running state of the fractional-order chemostat system when fractional order α at the Hopf bifurcation point is shown. The fractional-order chemostat system changed its stability once Hopf bifurcation occurred. Therefore, we conjecture that the system of fractional-order chemostat may be lost or gain its stability when the fractional order α is less than the Hopf bifurcation point, or $α < 0.9$ or otherwise. This shows that increasing or decreasing the value of α may destabilise the stable state of the chemostat system. Therefore, these results show that the running state of the fractional-order chemostat system is affected by the value of α.

3.4. Hopf bifurcation analysis of initial concentration of substrate

The initial concentration of the substrate, S₀, was chosen as the control parameter, while fractional order α was fixed. The solutions of steady-state of the fractional-order chemostat model with S₀ as the control parameter are

(i) Washout:

$(S_{0}^{*}, X_{0}^{*}) = (S_{0},0),$ (3.30)

(ii) No Washout:

$(S_{1}^{*}, X_{1}^{*}) = (\frac{1}{8}, \frac{533 (- 1 + 8 S_{0})}{6400}) .$ (3.31)

The eigenvalues obtained from the washout steady-state solution are

$λ_{1} = - \frac{1}{50},$ (3.32)

$λ_{2} = - \frac{7 (1 - 8 S_{0})}{50 (- 7 - 4 S_{0}},$ (3.33)

and the eigenvalues from the no-washout steady-state solution are

$λ_{1} = - 4204 + 1652 S_{0} + \frac{\sqrt{2} \sqrt{1364552 S_{0}^{2} - 245579768 S_{0} + 38666153}}{399750} i,$ (3.34)

$λ_{2} = - 4204 + 1652 S_{0} - \frac{\sqrt{2} \sqrt{1364552 S_{0}^{2} - 245579768 S_{0} + 38666153}}{399750} i .$ (3.35)

These satisfied the first condition of Hopf bifurcation in Theorem 3. There exist a pair of complex conjugate roots in terms of S₀, and the other eigenvalues were negative real roots in terms of S₀. The transversality condition as the third condition is also satisfied. According to Theorem 3, the critical value of the fractional order as stated in the second condition can be obtained as follows

$m (S_{0}^{*}) = \frac{α^{*} π}{2} - m i n | a r g (λ) | = 0.$ (3.36)

By referring to the study by [18], Eq. (3.37) can also be calculated as

$\frac{q^{'} (S_{0}^{*}) p (S_{0}^{*}) - q (S_{0}^{*}) p^{'} (S_{0}^{*})}{q^{2} (S_{0}^{*}) + p^{2} (S_{0}^{*})} \neq 0.$ (3.37)

From the calculations, the critical value of the initial concentration of the substrate is $S_{0} = 2.54$ . When $S_{0} = 2.54$ , the chemostat system shows Hopf bifurcation, at which the stability of the system would be altered.

Figure 4 presents the phase portrait diagrams of concentration of cell mass against concentration of substrate when $α = 1$ for different values of the initial concentration of the substrate, which are $S_{0} = 2$ , $S_{0} = 2.54$ and $S_{0} = 3.5$ .

Figure 4. Phase portrait plot of chemostat system with $α = 1$ (a) $S_{0} = 2$ (b) $S_{0} = 2.54$ and (c) $S_{0} = 3.5$ .

DownLoad: Full-Size Img PowerPoint

The change in the running state when the value of the initial substrate concentration passes through the Hopf bifurcation point is shown. The stability of the fractional-order chemostat system changed once Hopf bifurcation occurred. In Figure 3(a), the fractional-order chemostat system is in a stable state when the initial substrate concentration value is less than the Hopf bifurcation point, or $S_{0} < 2.54$ . Meanwhile, when the value of the initial substrate concentration passes through the Hopf bifurcation point, or $S_{0} \geq 2.54$ , the fractional-order chemostat system lost its stability. This shows that increasing the value of the initial substrate concentration may destabilise the stable state of the chemostat system. These results show that the running state of the fractional-order chemostat system is affected by the value of the initial substrate concentration. In real-life application, the value of the initial substrate should remain at $S_{0} \geq 2.54$ to ensure that the chemostat system is at the unstable state. This is because the unstable state is suitable for the production of cell mass [1]. Unstable state means the system always move away after small disturbance, so the system must be at the unstable state because there will be a change in amount of cell mass production.

Figure 5 depicts the phase portrait diagrams of cell mass concentration against substrate concentration when $α = 0.9$ for different values of the initial concentration of the substrate, which are $S_{0} = 2$ , $S_{0} = 2.54$ and $S_{0} = 3.5$ .

Figure 5. Phase portrait plot of chemostat system with $α = 0.9$ (a) $S_{0} = 2$ (b) $S_{0} = 2.54$ and (c) $S_{0} = 3.5$ .

DownLoad: Full-Size Img PowerPoint

The Hopf bifurcation points of system of fractional-order chemostat and system of integer-order chemostat are different. Figure 4 shows the fractional-order chemostat system at a stable state for all values of the initial substrate concentration when $α = 0.9$ . The chemostat system destabilised the stable state when the initial substrate concentration value is $S_{0} \geq 2.54$ , as shown in Figure 3(b) and Figure 3(c). This shows that the dynamic behaviour of the fractional-order chemostat system is different compared with the integer-order chemostat system. In actual application, the value of the initial substrate should remain at $S_{0} \geq 2.54$ to ensure that the chemostat system can be well controlled in order to be suitable for cell mass production.

4. Conclusions

The stability analysis of the fractional-order chemostat model was conducted based on the stability theory of FDEs. The integer-order chemostat model was extended to the FDEs. There are two steady-state solutions obtained, which are washout and no-washout steady-state solutions. The Hopf bifurcation of the order of α occured at the solutions of steady-state when the Hopf bifurcation conditions is fulfilled. The results show that the increasing or decreasing the value of α may stabilise the unstable state of the chemostat system. Therefore, the running state of the fractional-order chemostat system is affected by the value of α. The Hopf bifurcation of the initial concentration of the substrate, S₀, also occurred when the Hopf bifurcation condition is fulfilled. As the evidence from the phase portrait plots, increase the value of the initial substrate concentration may destabilise the stable state of the chemostat system. The value of the initial substrate should remain at $S_{0} \geq 2.54$ to ensure that the chemostat system is at the unstable state since the unstable state is suitable for the production of cell mass. These dynamical analyses are important to provide suitable values of the fractional-order and the parameters in order to ensure the controllability and stability of the chemostat to suit the actual chemostat environment.

References

[1]	L. Zhao, J. Song, Y. Liu, C. Song, C. Yi, Mapping the epigenetic modifications of DNA and RNA, Protein Cell, 11 (2020), 792–808. https://doi.org/10.1007/s13238-020-00733-7 doi: 10.1007/s13238-020-00733-7
[2]	L. D. Moore, T. Le, G. Fan, DNA methylation and its basic function, Neuropsychopharmacology, 38 (2013), 23–38. https://doi.org/10.1038/npp.2012.112 doi: 10.1038/npp.2012.112
[3]	N. Zhang, C. Lin, X. Huang, A. Kolbanovskiy, B. E. Hingerty, S. Amin, et al., Methylation of cytosine at C5 in a CpG sequence context causes a conformational switch of a benzo[a]pyrene diol epoxide-N²-guanine adduct in DNA from a minor groove alignment to intercalation with base displacement, J. Mol. Biol., 346 (2005), 951–965. https://doi.org/10.1016/j.jmb.2004.12.027 doi: 10.1016/j.jmb.2004.12.027
[4]	A. Breiling, F. Lyko, Epigenetic regulatory functions of DNA modifications: 5-methylcytosine and beyond, Epigenet. Chromatin, 8 (2015), 24. https://doi.org/10.1186/s13072-015-0016-6 doi: 10.1186/s13072-015-0016-6
[5]	A. Jeltsch, R. Z. Jurkowska, New concepts in DNA methylation, Trends Biochem. Sci., 39 (2014), 310–318. https://doi.org/10.1016/j.tibs.2014.05.002 doi: 10.1016/j.tibs.2014.05.002
[6]	D. Schübeler, Function and information content of DNA methylation, Nature, 517 (2015), 321–326. https://doi.org/10.1038/nature14192 doi: 10.1038/nature14192
[7]	N. P. Blackledge, R. Klose, CpG island chromatin, Epigenetics, 6 (2011), 147–152. https://doi.org/10.4161/epi.6.2.13640 doi: 10.4161/epi.6.2.13640
[8]	F. J. Clasen, R. E. Pierneef, B. Slippers, O. Reva, EuGI: a novel resource for studying genomic islands to facilitate horizontal gene transfer detection in eukaryotes, BMC Genomics, 19 (2018). https://doi.org/10.1186/s12864-018-4724-8 doi: 10.1186/s12864-018-4724-8
[9]	X. Guo, Y. Guo, H. Chen, X. Liu, P. He, W. Li, et al., Systematic comparison of genome information processing and boundary recognition tools used for genomic island detection, Comput. Biol. Med., 166 (2023), 107550. https://doi.org/10.1016/j.compbiomed.2023.107550 doi: 10.1016/j.compbiomed.2023.107550
[10]	Q. Dai, C. Bao, Y. Hai, S. Ma, T. Zhou, C. Wang, et al., MTGIpick allows robust identification of genomic islands from a single genome, Briefings Bioinf., 19 (2018), 361–373. https://doi.org/10.1093/bib/bbw118 doi: 10.1093/bib/bbw118
[11]	A. Chialastri, S. Sarkar, E. E. Schauer, S. Lamba, S. S. Dey, Combinatorial quantification of 5mC and 5hmC at individual CpG dyads and the transcriptome in single cells reveals modulators of DNA methylation maintenance fidelity, preprint.
[12]	G. Luo, M. A. Blanco, E. L. Greer, C. He, Y. Shi, DNA N⁶-methyladenine: a new epigenetic mark in eukaryotes?, Nat. Rev. Mol. Cell Biol., 16 (2015), 705–710. https://doi.org/10.1038/nrm4076 doi: 10.1038/nrm4076
[13]	J. Beaulaurier, E. E. Schadt, G. Fang, Deciphering bacterial epigenomes using modern sequencing technologies, Nat. Rev. Genet., 20 (2019), 157–172. https://doi.org/10.1038/s41576-018-0081-3 doi: 10.1038/s41576-018-0081-3
[14]	M. Ehrlich, G. G. Wilson, K. C. Kuo, C. W. Gehrke, N4-methylcytosine as a minor base in bacterial DNA, Journal of Bacteriology, 169 (1987), 939–943. https://doi.org/10.1128/jb.169.3.939-943.1987 doi: 10.1128/jb.169.3.939-943.1987
[15]	F. Rodriguez, I. A. Yushenova, D. DiCorpo, I. R. Arkhipova, Bacterial N4-methylcytosine as an epigenetic mark in eukaryotic DNA, Nat. Commun., 13 (2022), 1072. https://doi.org/10.1038/s41467-022-28471-w doi: 10.1038/s41467-022-28471-w
[16]	M. Yu, L. Ji, D. A. Neumann, D. Chung, J. Groom, J. Westpheling, et al., Base-resolution detection of N 4-methylcytosine in genomic DNA using 4mC-Tet-assisted-bisulfite- sequencing, Nucleic Acids Res., 43 (2015), 148. https://doi.org/10.1093/nar/gkv738 doi: 10.1093/nar/gkv738
[17]	S. Ardui, A. Ameur, J. R. Vermeesch, M. S. Hestand, Single molecule real-time (SMRT) sequencing comes of age: applications and utilities for medical diagnostics, Nucleic Acids Res., 46 (2018), 2159–2168. https://doi.org/10.1093/nar/gky066 doi: 10.1093/nar/gky066
[18]	B. Manavalan, S. Basith, T. H. Shin, D. Y. Lee, L. Wei, G. Lee, 4mCpred-EL: An ensemble learning framework for identification of DNA N4-methylcytosine sites in the mouse genome, Cells, 8 (2019), 1332. https://doi.org/10.3390/cells8111332 doi: 10.3390/cells8111332
[19]	W. He, C. Jia, Q. Zou, 4mCPred: machine learning methods for DNA N⁴-methylcytosine sites prediction, Bioinformatics, 35 (2019), 593–601. https://doi.org/10.1093/bioinformatics/bty668 doi: 10.1093/bioinformatics/bty668
[20]	M. M. Hasan, B. Manavalan, W. Shoombuatong, M. S. Khatun, H. Kurata, i4mC-Mouse: Improved identification of DNA N4-methylcytosine sites in the mouse genome using multiple encoding schemes, Comput. Struct. Biotechnol. J., 18 (2020), 906–912. https://doi.org/10.1016/j.csbj.2020.04.001 doi: 10.1016/j.csbj.2020.04.001
[21]	W. Chen, H. Yang, P. Feng, H. Ding, H. Lin, iDNA4mC: identifying DNA N⁴-methylcytosine sites based on nucleotide chemical properties, Bioinformatics, 33 (2017), 3518–3523. https://doi.org/10.1093/bioinformatics/btx479 doi: 10.1093/bioinformatics/btx479
[22]	B. Manavalan, S. Basith, T. H. Shin, L. Wei, G. Lee, Meta-4mCpred: A sequence-based meta-predictor for accurate DNA 4mC site prediction using effective feature representation, Mol. Ther.-Nucleic Acids, 16 (2019), 733–744. https://doi.org/10.1016/j.omtn.2019.04.019 doi: 10.1016/j.omtn.2019.04.019
[23]	H. Xu, P. Jia, Z. Zhao, Deep4mC: systematic assessment and computational prediction for DNA N4-methylcytosine sites by deep learning, Briefings Bioinf., 22 (2021). https://doi.org/10.1093/bib/bbaa099 doi: 10.1093/bib/bbaa099
[24]	Q. Liu, J. Chen, Y. Wang, S. Li, C. Jia, J. Song, et al., DeepTorrent: a deep learning-based approach for predicting DNA N4-methylcytosine sites, Briefings Bioinf., 22 (2021). https://doi.org/10.1093/bib/bbaa124 doi: 10.1093/bib/bbaa124
[25]	X. Yu, J. Ren, Y. Cui, R. Zeng, H. Long, C. Ma, DRSN4mCPred: accurately predicting sites of DNA N4-methylcytosine using deep residual shrinkage network for diagnosis and treatment of gastrointestinal cancer in the precision medicine era, Front. Med., 10 (2023). https://doi.org/10.3389/fmed.2023.1187430 doi: 10.3389/fmed.2023.1187430
[26]	Y. Yu, W. He, J. Jin, G. Xiao, L. Cui, R. Zeng, et al., iDNA-ABT: advanced deep learning model for detecting DNA methylation with adaptive features and transductive information maximization, Bioinformatics, 37 (2021), 4603–4610. https://doi.org/10.1093/bioinformatics/btab677 doi: 10.1093/bioinformatics/btab677
[27]	J. Jin, Y. Yu, R. Wang, X. Zeng, C. Pang, Y. Jiang, et al., iDNA-ABF: multi-scale deep biological language learning model for the interpretable prediction of DNA methylations, Genome Biol., 23 (2022). https://doi.org/10.1186/s13059-022-02780-1 doi: 10.1186/s13059-022-02780-1
[28]	W. Zeng, A. Gautam, D. H. Huson, MuLan-Methyl-multiple transformer-based language models for accurate DNA methylation prediction, GigaScience, 12 (2023). https://doi.org/10.1093/gigascience/giad054 doi: 10.1093/gigascience/giad054
[29]	M. U. Rehman, H. Tayara, K. T. Chong, DCNN-4mC: Densely connected neural network based N4-methylcytosine site prediction in multiple species, Comput. Struct. Biotechnol. J., 19 (2021), 6009–6019. https://doi.org/10.1016/j.csbj.2021.10.034 doi: 10.1016/j.csbj.2021.10.034
[30]	S. Park, M. U. Rehman, F. Ullah, H. Tayara, K. T. Chong, I. Birol, iCpG-Pos: an accurate computational approach for identification of CpG sites using positional features on single-cell whole genome sequence data, Bioinformatics, 39 (2023). https://doi.org/10.1093/bioinformatics/btad474 doi: 10.1093/bioinformatics/btad474
[31]	L. Zhang, X. Xiao, Z. Xu, iPromoter-5mC: A novel fusion decision predictor for the identification of 5-Methylcytosine sites in genome-Wide DNA promoters, Front. Cell Dev. Biol., 8 (2020). https://doi.org/10.3389/fcell.2020.00614 doi: 10.3389/fcell.2020.00614
[32]	D. Y. Lim, M. U. Rehman, K. T. Chong, iRG-4mC: Neural network based tool for identification of DNA 4mC sites in rosaceae genome, Symmetry, 13 (2021), 899. https://doi.org/10.3390/sym13050899 doi: 10.3390/sym13050899
[33]	M. U. Rehman, K. J. Hong, H. Tayara, K. T. Chong, m6A-NeuralTool: Convolution neural tool for RNA N6-Methyladenosine site identification in different species, IEEE Access, 9 (2021), 17779–17786. https://doi.org/10.1109/access.2021.3054361 doi: 10.1109/access.2021.3054361
[34]	Q. H. Nguyen, H. V. Tran, B. P. Nguyen, T. T. T. Do, Identifying transcription factors that prefer binding to methylated DNA using reduced G-Gap dipeptide composition, ACS Omega, 7 (2022), 32322–32330. https://doi.org/10.1021/acsomega.2c03696 doi: 10.1021/acsomega.2c03696
[35]	Z. Li, H. Jiang, L. Kong, Y. Chen, K. Lang, X. Fan, et al., Deep6mA: A deep learning framework for exploring similar patterns in DNA N6-methyladenine sites across different species, PLOS Comput. Biol., 17 (2021), 1008767. https://doi.org/10.1371/journal.pcbi.1008767 doi: 10.1371/journal.pcbi.1008767
[36]	X. Cheng, J. Wang, Q. Li, T. Liu, BiLSTM-5mC: A bidirectional long short-term memory-based approach for predicting 5-Methylcytosine sites in genome-wide DNA promoters, Molecules, 26 (2021), 7414. https://doi.org/10.3390/molecules26247414 doi: 10.3390/molecules26247414
[37]	Z. Abbas, H. Tayara, K. T. Chong, 4mCPred-CNN-Prediction of DNA N4-Methylcytosine in the mouse genome using a convolutional neural network, Genes, 12 (2021), 296. https://doi.org/10.3390/genes12020296 doi: 10.3390/genes12020296
[38]	J. Jin, Y. Yu, L. Wei, Mouse4mC-BGRU: Deep learning for predicting DNA N4-methylcytosine sites in mouse genome, Methods, 204 (2022), 258–262. https://doi.org/10.1016/j.ymeth.2022.01.009 doi: 10.1016/j.ymeth.2022.01.009
[39]	P. Zheng, G. Zhang, Y. Liu, G. Huang, MultiScale-CNN-4mCPred: a multi-scale CNN and adaptive embedding-based method for mouse genome DNA N4-methylcytosine prediction, BMC Bioinformatics, 24 (2023). https://doi.org/10.1186/s12859-023-05135-0 doi: 10.1186/s12859-023-05135-0
[40]	T. Nguyen-Vo, Q. H. Trinh, L. Nguyen, P. Nguyen-Hoang, S. Rahardja, B. P. Nguyen, i4mC-GRU: Identifying DNA N4-Methylcytosine sites in mouse genomes using bidirectional gated recurrent unit and sequence-embedded features, Comput. Struct. Biotechnol. J., 21 (2023), 3045–3053. https://doi.org/10.1016/j.csbj.2023.05.014 doi: 10.1016/j.csbj.2023.05.014
[41]	C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, et al., Going deeper with convolutions, in 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, (2015), 1–9. https://doi.org/10.1109/CVPR.2015.7298594
[42]	X. Li, W. Wang, X. Hu, J. Yang, Selective kernel networks, in 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), (2019), 510–519. https://doi.org/10.1109/CVPR.2019.00060
[43]	S. Woo, S. Debnath, R. Hu, X. Chen, Z. Liu, I. S. Kweon, et al., ConvNeXt V2: Co-Designing and scaling ConvNets with masked autoencoders, in 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), (2023), 16133–16142. https://doi.org/10.1109/CVPR52729.2023.01548
[44]	A. G. Howard, M. Zhu, B. Chen, D. Kalenichenko, W. Wang, T. Weyand, et al., MobileNets: efficient convolutional neural networks for mobile vision applications, preprint, arXiv: 1704.04861.
[45]	P. Ye, Y. Luan, K. Chen, Y. Liu, C. Xiao, Z. Xie, MethSMRT: an integrative database for DNA N6-methyladenine and N4-methylcytosine generated by single-molecular real-time sequencing, Nucleic Acids Res., 45 (2017), 85–89. https://doi.org/10.1093/nar/gkw950 doi: 10.1093/nar/gkw950
[46]	L. Fu, B. Niu, Z. Zhu, S. Wu, W. Li, CD-HIT: accelerated for clustering the next-generation sequencing data, Bioinformatics, 28 (2012), 3150–3152. https://doi.org/10.1093/bioinformatics/bts565 doi: 10.1093/bioinformatics/bts565
[47]	W. Chen, H. Tang, J. Ye, H. Lin, K. Chou, iRNA-PseU: Identifying RNA pseudouridine sites, Mol. Ther.-Nucleic Acids, 5 (2016), 332.
[48]	T. Nguyen-Vo, Q. H. Nguyen, T. T. T. Do, T. Nguyen, S. Rahardja, B. P. Nguyen, iPseU-NCP: Identifying RNA pseudouridine sites using random forest and NCP-encoded features, BMC Genomics, 20 (2019). https://doi.org/10.1186/s12864-019-6357-y doi: 10.1186/s12864-019-6357-y
[49]	A. S. Nair, S. P. Sreenadhan, A coding measure scheme employing electron-ion interaction pseudopotential (EⅡP), Bioinformation, 1 (2006), 197–202.
[50]	R. Avenash, P. Viswanath, Semantic segmentation of satellite images using a modified CNN with hard-swish activation function, in Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, (2019), 413–420.
[51]	K. He, X. Zhang, S. Ren, J. Sun, Deep residual learning for image recognition, in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), (2016), 770–778. https://doi.org/10.1109/CVPR.2016.90
[52]	A. Krizhevsky, I. Sutskever, G. E. Hinton, ImageNet classification with deep convolutional neural networks, Commun. ACM, 60 (2017), 84–90. https://doi.org/10.1145/3065386 doi: 10.1145/3065386
[53]	S. Ioffe, C. Szegedy, Batch normalization: Accelerating deep network training by reducing internal covariate shift, preprint, arXiv: 1502.03167.
[54]	J. L. Ba, J. R. Kiros, G. E. Hinton, Layer normalization, preprint, arXiv: 1607.06450.
[55]	M. Sokolova, G. Lapalme, A systematic analysis of performance measures for classification tasks, Inf. Process. Manage., 45 (2009), 427–437. https://doi.org/10.1016/j.ipm.2009.03.002 doi: 10.1016/j.ipm.2009.03.002
[56]	H. Lv, F. Dao, D. Zhang, Z. Guan, H. Yang, W. Su, et al., iDNA-MS: An integrated computational tool for detecting DNA modification sites in multiple genomes, iScience, 23 (2020), 100991. https://doi.org/10.1016/j.isci.2020.100991 doi: 10.1016/j.isci.2020.100991
[57]	A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, et al., Attention is all you need, preprint, arXiv: 1706.03762v5.

This article has been cited by:

Xiaomeng Ma, Zhanbing Bai, Sujing Sun, Stability and bifurcation control for a fractional-order chemostat model with time delays and incommensurate orders, 2022, 20, 1551-0018, 437, 10.3934/mbe.2023020

Reader Comments

Your name:*

Email:*
© 2024 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)