A method for monitoring the dynamic growth and development of plants

Abstract

The application discloses a kind of plant growth dynamic monitoring methods, comprising: obtaining two groups of single-cell transcriptome data, wherein the first group of data is control group, the second group of data is change group, corresponding to different development stages respectively;By singular value analysis, the virtual cell state matrix of the second group of data in the development stage corresponding to the control group is constructed;The state change response characteristics of the single cell corresponding to the second group of data between different development stages are obtained;Single cell is clustered using state change response characteristics.The application solves the problem that single-cell technology destroys sequencing cells and therefore cannot track the state change of each cell over a long period of time during plant growth and development, enabling the application to uncover dynamic differentiation characteristics of cells during plant growth that cannot be discovered by existing methods.

Description

Technical Field

[0001] This invention relates to the field of plant technology, and in particular to a method for dynamic monitoring of plant growth and development. Background Technology

[0002] In recent years, the development of single-cell sequencing technology has made it a highly sought-after technology, and the increasingly mature industry chain has led to a surge in related research projects and results. Compared to traditional bulk sequencing, single-cell transcriptome (scRNA-seq) sequencing technology can provide researchers with information on gene expression in individual cells. In 2019, researchers first applied scRNA-seq technology to the study of root tissue development in Arabidopsis thaliana. This study identified the main expression characteristics of cell types such as phloem, column cells, and QCs, and further revealed the complex process of root cell differentiation in Arabidopsis thaliana and key transcription factors such as LAR3, ATHB-20, and GATA4 through pseudo-time series analysis. Subsequently, more research on plant single cells has emerged both domestically and internationally. These studies currently mainly focus on cell differentiation and tissue development in model plants such as Arabidopsis thaliana, tomato, and rice.

[0003] Research on single-cell organisms in plants is providing researchers with entirely new ideas and perspectives. For example, in a study on Arabidopsis root tip development using single-cell technology, researchers mapped the cellular atlas of Arabidopsis root tips. They identified various cell populations, including root tip stem cells, cambium cells, and xylem cells, and used pseudo-time series analysis to depict their developmental trajectories and corresponding changes in the expression of differentiation-related genes. The results indicate that Arabidopsis root tips contain highly heterogeneous cell populations, and even within the same cell type, there are significant differences in gene modalities.

[0004] Studies on plant growth and development typically involve collecting samples at multiple growth stages and time points. The impact of changing conditions on cells is then inferred by comparing gene expression profiles under different conditions. Currently, single-cell data analysis of plant growth and development primarily relies on static gene expression profiles—measured values—to infer dynamic processes. However, current single-cell sequencing technologies lyse and disrupt cell characteristics during sequencing, making it impossible to pair cells from different growth stages. This random heterogeneity between cells can mask much of the information relevant to plant growth and development, and traditional methods based on static cell characteristics cannot uncover these signals. This results in a large amount of hidden information related to plant growth and development being overlooked.

[0005] Because single-cell sequencing technology lyses and damages cells during sequencing to acquire data, which is an irreversible process, it is impossible to obtain single-cell RNA-seq data of the same cell at different stages through biological experiments. Consequently, it is impossible to perform biological data analysis and comparison of the same sample at different stages.

[0006] Therefore, those skilled in the art are dedicated to developing a method for dynamic monitoring of plant growth and development. By creating real-virtual cell pairs of cell mapping, the difficulty of comparing RNA-seq data of the same cell at different stages has been solved, and this method has been successfully applied to plant cell growth and development research. Summary of the Invention

[0007] To achieve the above objectives, the present invention provides a method for monitoring plant growth dynamics, comprising the following steps:

[0008] Two sets of single-cell transcriptome data were obtained, where the first set of data X is the cell state matrix of the control group and the second set of data Y is the cell state matrix of the change group. The control group and the change group correspond to different developmental stages.

[0009] By using singular value analysis, a virtual cell state matrix of the second set of data Y corresponding to the developmental stage in the control group is constructed.

[0010] The state change response characteristics of single cells corresponding to the second set of data Y between the developmental stage corresponding to the control group and the developmental stage corresponding to the change group were obtained.

[0011] The single cells corresponding to the second set of data Y are clustered using the state change response features.

[0012] Furthermore, the developmental stages include: phloem tissue growth period, plant stem cell growth period, root cap growth period, and xylem growth period.

[0013] Furthermore, the control group corresponds to the growth period of the phloem tissue, and the change group corresponds to the growth period of the plant stem cells.

[0014] Furthermore, the state change response characteristics are obtained using the difference between the virtual cell state matrix and the second set of data Y.

[0015] Furthermore, using singular value decomposition, the covariance matrix of the single-cell data at different developmental stages is decomposed to obtain the component that maximizes the covariance between the single-cell data, and the virtual cell state matrix is ​​constructed based on this component.

[0016] Furthermore, the relationship between the second set of data Y and the virtual cell state matrix is ​​expressed by the following formula:

[0017]

[0018] in, V represents the virtual cell state matrix; V represents the state change response feature matrix.

[0019] Furthermore, the virtual cell state matrix is ​​represented by the following Formula 2:

[0020]

[0021] Where P represents the load matrix, T represents the transpose matrix, and S′ represents the singular value decomposition matrix.

[0022] Furthermore, the load matrix is ​​obtained by maximizing the following Equation 3 using the NIPALS method:

[0023]

[0024] Furthermore, the process of solving the state change response feature matrix is ​​as follows:

[0025] because It can be obtained Due to V and Irrelevant, P T (P T P) -1 PV≈0, It can be approximated as P T (P T P) -1 PU, V can be calculated as

[0026] V = YP T (P T P) -1 PY

[0027] Given Y, V is estimated using P.

[0028] Furthermore, the process of solving the state change response feature matrix is ​​as follows:

[0029] Assumption X and X are different cells from the same feature space, and X replaces the first one in Formula 3. Replace P with C to represent If there is a possible mismatch between X and E, then Formula 3 is approximately equivalent to Formula 4:

[0030]

[0031] Since X and V are uncorrelated, for any P and C, Cov(XC,VP)≈0, then Formula 4 is expressed as Formula 5:

[0032]

[0033] Given X and Y, P is calculated based on Formula 5, and then... V is then calculated based on the aforementioned formula.

[0034] This invention offers the following technical advantages: It is applicable to pairwise RNA-seq studies comparing samples from different stages of plant growth and development. It allows for dynamic monitoring and comparison of each cell state across different stages of plant growth and development (e.g., from the phloem growth stage, plant stem cell growth stage, root cap growth stage, xylem growth stage, etc.). This invention solves the problem that single-cell technology destroys sequencing cells, making it impossible to track changes in the state of each cell over long periods during plant growth and development. Therefore, this invention can uncover dynamic differentiation characteristics of cells during plant growth that are undetectable by existing methods.

[0035] The following will further explain the concept, specific structure, and technical effects of the present invention in conjunction with the accompanying drawings, so as to fully understand the purpose, features, and effects of the present invention. Attached Figure Description

[0036] Figure 1 This is a flowchart of a preferred embodiment of the present invention;

[0037] Figure 2 This is an experimental result diagram of a preferred embodiment of the present invention. Detailed Implementation

[0038] The following description, with reference to the accompanying drawings, illustrates several preferred embodiments of the present invention to make its technical content clearer and easier to understand. The present invention can be embodied in many different forms, and the scope of protection of the present invention is not limited to the embodiments mentioned herein.

[0039] In the accompanying drawings, components with the same structure are indicated by the same numerical designation, and components with similar structures or functions are indicated by similar numerical designations. The dimensions and thicknesses of each component shown in the drawings are arbitrary, and the present invention does not limit the dimensions and thicknesses of each component. To make the illustrations clearer, the thickness of some components has been appropriately exaggerated in the drawings.

[0040] This invention provides a method for dynamic monitoring of plant growth and development, suitable for pairwise comparisons of samples at different stages of plant growth and development in single-cell RNA-seq studies. The method dynamically monitors and compares the state of each cell as the plant grows and develops. These different stages include, but are not limited to, the phloem growth phase, plant stem cell growth phase, root and cap growth phase, and xylem growth phase.

[0041] In this invention, the change of the same cell between state 1 (stage 1) and state 2 (stage 2) is represented by the difference between mapped cells in another state space that corresponds to it one-to-one. That is, for each real cell (in state 1 or state 2), a virtual cell in the corresponding other state space is constructed using singular value decomposition (SVD). The difference between the real and virtual cell pairs represents the cell's response characteristics to state changes. Specifically, SVD is used to decompose the covariance matrix of single-cell data at different growth stages, obtaining the component that maximizes the covariance between the two groups of cells, and virtual cells are constructed based on this component. When comparing the differences between real and virtual cell pairs, variables related to the cell state dynamics process under different plant growth states are retained, while irrelevant information is removed. Then, cell clustering is performed based on the dynamic state variables of plant growth and development. Specifically, as shown... Figure 1 As shown, the method of the present invention includes:

[0042] Step 1: Obtain two sets of single-cell transcriptome data, corresponding to two developmental stages. The first set of data is the cell state matrix of the control group, and the second set is the cell state matrix of the variation group. The control group and the variation group each correspond to two developmental stages, which can be selected according to actual needs. For example, the control group could select the phloem tissue growth stage, and the variation group could select the plant stem cell growth stage. Let X represent the cell state matrix of the control group, and Y represent the cell state matrix of the variation group.

[0043] Step Two: Through singular value analysis, construct virtual cells of the variant group within the control group, i.e., the virtual cell state matrix of the variant group at the developmental stage of the control group. Considering that X represents the baseline, Y can be decomposed into:

[0044]

[0045] Among them, matrix V is the virtual cell state matrix of Y in the cell feature space of the control group. Therefore, it has the same dimension as the Y matrix, that is, it has a one-to-one correspondence of cell number and gene number. V represents the state change response characteristics of the cells contained in Y from the first developmental stage stage 1 (the developmental stage corresponding to the control group) to the second developmental stage stage 2 (the developmental stage corresponding to the change group). This change is mainly caused by the growth and developmental changes between stage 2 and stage 1. That is, V represents the difference between the real and virtual cell pairs, representing the response characteristics of the cell to the state change.

[0046] By solving the virtual cell state matrix Using Y and The difference can be used to obtain the state change response characteristic V.

[0047] Step 3: By performing cluster analysis on cells based on the state change response characteristics V, the state dynamics of single cells at different developmental stages can be monitored.

[0048] For the virtual cell state matrix, singular value decomposition is used to decompose the covariance matrix of single-cell data at different developmental stages, obtaining the component that maximizes the covariance between single-cell data. Based on this component, the virtual cell state matrix can be constructed.

[0049] Assume the state matrix of the virtual cell This can be characterized using the singular value decomposition method: Where P represents the load matrix. T Let S' denote the transpose matrix, and S′ denote the score matrix. The load matrix P can be obtained by maximizing the following formula using the NIPALS method:

[0050]

[0051] The NIPALS method can be obtained from the existing technical literature Alin, A., Comparison of PLS ​​algorithms when number of objects is much larger than number of variables. Statistical Papers, 2009, 50(4): p.711-720.

[0052] In some implementations, the process of solving for V is as follows: Since It can be obtained Due to V and Irrelevant, P T (P T P) -1 PV≈0, It can be approximated as P T (P T P) -1 PY, V can be calculated as

[0053] V = YP T (P T P) -1 PY (3)

[0054] Equation (3) shows that, given Y, V can be estimated by P.

[0055] In some implementations, the process of solving for V is as follows: assuming X and X are different cells from the same feature space, and X replaces the first one in equation (2). Replace P with C to represent The possible mismatch between X and E can be approximated by equation (2) as follows:

[0056]

[0057] Since X and V are uncorrelated, for any P and C, Cov(XC,VP)≈0, and formula (4) can be expressed as:

[0058]

[0059] Formulas (4) and (5) simplify the calculation process of P. Given X and Y, P can be calculated, and thus the result can be obtained. V is calculated using formula (1).

[0060] like Figure 2 As shown, the left image is the result obtained after processing cell data using the method of this invention. Because each cell is described using its dynamic characteristics, the differences between different cell types are extracted and amplified. The right image, however, is the result obtained using traditional classification methods. Since the static characteristics of the raw cell sequencing values ​​are directly used for description, intercellular heterogeneity often fails to display meaningful clustering features due to noise interference. These two images demonstrate that the method provided by this invention solves the problem that single-cell technology destroys sequenced cells, making it impossible to track the state changes of each cell over a long period during plant growth and development. It can uncover dynamic differentiation characteristics of cells during plant growth that existing methods cannot detect.

[0061] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A method for dynamic monitoring of plant growth and development, characterized in that, Includes the following steps: Two sets of single-cell transcriptome data were obtained, where the first set of data X is the cell state matrix of the control group and the second set of data Y is the cell state matrix of the change group. The control group and the change group correspond to different developmental stages. Using singular value decomposition, the covariance matrix of single-cell data at different developmental stages is decomposed to obtain the component that maximizes the covariance among the single-cell data. Based on this component, a virtual cell state matrix of the second group of data Y at the developmental stage corresponding to the control group is constructed. The virtual cell state matrix is ​​the virtual cell state matrix of Y in the cell feature space of the control group, and it is completely consistent with the number of cells and genes of Y. Based on relational The state change response characteristics of single cells corresponding to the second set of data Y between the developmental stages corresponding to the control group and the developmental stages corresponding to the change group were obtained. ;in, express The specific state change response characteristics of the cells contained in the control group from the developmental stage to the developmental stage of the modified group; Utilizing the state change response characteristics Cluster the single cells corresponding to the second set of data Y.

2. The method for dynamic monitoring of plant growth and development as described in claim 1, characterized in that, The developmental stages include: phloem tissue growth period, plant stem cell growth period, root and crown growth period, and xylem growth period.

3. The method for dynamic monitoring of plant growth and development as described in claim 2, characterized in that, The control group corresponds to the growth period of the phloem tissue, and the change group corresponds to the growth period of the plant stem cells.

4. The method for dynamic monitoring of plant growth and development as described in claim 1, characterized in that, The state change response feature is obtained by using the difference between the virtual cell state matrix and the second set of data Y.

5. The method for dynamic monitoring of plant growth and development as described in claim 4, characterized in that, The relationship between the second set of data Y and the virtual cell state matrix is ​​expressed by the following formula: in, V represents the virtual cell state matrix; V represents the state change response feature matrix.

6. The method for dynamic monitoring of plant growth and development as described in claim 5, characterized in that, The virtual cell state matrix is ​​represented by the following formula: Where P represents the load matrix and T represents the transpose matrix. This represents the singular value decomposition matrix.

7. The method for dynamic monitoring of plant growth and development as described in claim 6, characterized in that, The load matrix is ​​obtained by maximizing the following Equation 3 using the NIPALLS method: 。 8. The method for dynamic monitoring of plant growth and development as described in claim 7, characterized in that, The process of solving the state change response feature matrix is ​​as follows: because ,get ; due to V and Unrelated , Approximately , V Calculated as Given Y, we obtain the following through P estimation: V .

9. The method for dynamic monitoring of plant growth and development as described in claim 7, characterized in that, The process of solving the state change response feature matrix is ​​as follows: Assumption X and X are different cells from the same feature space, and X replaces the first one in Formula 3. Replace P with C to represent If there is a possible mismatch between X and E, then Formula 3 is approximately equivalent to Formula 4: Since X and V are uncorrelated, for any P and C, Then, Formula 4 is expressed as Formula 5: Given X and Y, P is calculated based on Formula 5, and then... V is calculated based on Formula 1.

Images