Method for constructing soil factor microbial diversity database based on literature

By constructing a literature-based Access database, the problem of collecting and analyzing soil factors and microbial diversity data is solved, timeliness and multiple factor analysis is achieved globally, machine learning research is supported, and a database that is easy to query and expand is provided. The problems of cumbersome research and insufficient data timeliness in the prior art have been solved.

CN114661712BActive Publication Date: 2024-11-29NANKAI UNIV
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202210249018.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-03-14
Publication Date
2024-11-29
Estimated Expiration
2042-03-14

AI Technical Summary

Technical Problem

The lack of a method for efficiently collecting, storing and analyzing data on the relationship between soil factors and microbial diversity across the globe, making research cumbersome and difficult to achieve timeliness and comprehensive data support.

Method used

By collecting global soil factor-microbial literature, extracting and organizing relevant data, using Access database for storage, updating, visualizing and analyzing, and developing dynamic charts in combination with SQL, VBA and controls, a literature-based soil factor microbial diversity database is constructed.

Benefits of technology

It provides a global multifactor analysis database that is time-efficient and scalable, supports subsequent artificial intelligence research, and explores the impact of soil factors on microbial diversity through machine learning algorithms.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114661712B_ABST
    Figure CN114661712B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for constructing a soil factor microbial diversity database based on literature. First, the key indicator parameters in the field of soil and soil microorganisms are studied; then the indicators are extracted, including article publication information, research context and methods, soil microbial community indicators and basic database information of soil indicators; secondly, data screening principles are specified to collect data from massive soil microbial historical documents, store them into an Access database, and use the Access database to update, visualize and analyze the data. The present invention constructs a soil microbial database based on literature, which can provide users with convenient documents and data for query, especially for other research that may be conducted using artificial intelligence in the future.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the field of database construction, and in particular relates to a method for constructing a soil factor microbial diversity database based on literature. Background Art

[0002] Soil microorganisms are huge in number, diverse in variety and active in function. They exist in the soil in the form of communities, affecting soil nutrient content and organic matter transformation, thereby affecting soil quality and plant growth. In recent years, soil microbial diversity has been severely damaged due to factors such as global climate change, overexploitation of land resources, and increasingly severe environmental pollution. Therefore, conducting research on the relationship between soil factors and soil microbial diversity can not only predict and evaluate the changes in microbial diversity and composition of soils contaminated by humans, but also further explore the functions of dominant microbial populations, providing a theoretical basis for land pollution control, protection and utilization, and sustainable development of soil.

[0003] Most of the existing literature is limited to the study of soil microbial diversity in a certain region or a single factor. There is no publicly recognized data set or public database. In addition to field measurements, scholars can only obtain data more comprehensively by reading massive historical documents. However, for authors, they need to read a large amount of literature for statistical collation, and they need to update every few years to ensure the real-time statistical knowledge of such literature. The workload is large and cumbersome. For readers, if they want to quickly obtain such statistical knowledge, they need to search for review literature and judge their authority. The knowledge obtained is also limited by what the literature writers want to show, such as region, year, etc. Moreover, the relationship between soil factors, pollution conditions and soil microbial diversity has not yet been explored. If you want to find the rules behind it, you need a lot of data support.

[0004] In addition, the rapid development of artificial intelligence in recent years has provided new ideas for data mining in the context of big data. Machine learning, as the core method of artificial intelligence, has been successfully applied to the field of ecological environment. This provides a new idea for studying the diversity and composition of microorganisms in any soil area, avoiding the tedious statistical work of later scholars, and is of great significance to the study of soil microorganisms. If we want to apply machine learning to explore the relationship between environmental pollution, soil environmental factors and soil microbial diversity and composition, screen and quantify the impact of important factors on soil microorganisms, and predict microbial diversity at a global scale, we must establish a complete and comprehensive soil factor-microbial diversity database.

[0005] In other fields such as clinical medicine and socialized government affairs, the idea and application of structured analysis of literature have emerged. In the field of clinical medicine, systematic reviews and meta-analyses have emerged. In the field of socialized government affairs, a scholar proposed a structured literature analysis framework in 2015 to summarize and analyze the existing foreign research literature on socialized government affairs, and has made certain progress, providing a certain basis and inspiration for the research on socialized government affairs. It can be seen that although there have been some preliminary studies in other fields, there are few studies in the field of soil microorganisms. This is because soil microorganism literature data has a strong timeliness, and the data in the literature is mostly in the form of charts and graphs, which is difficult to collect through intelligent methods such as literature language processing. Therefore, in this invention, the project members mainly read and analyze the literature manually, and build a database with timeliness according to appropriate data screening principles, which can be directly applied to subsequent artificial intelligence or other research. Summary of the invention

[0006] The problem to be solved by the present invention is to provide a method for constructing a soil factor microbial diversity database based on literature, which can collect data scattered in a large amount of soil microbial historical literature, store them into an Access database, and use the Access database to update, visualize and analyze the data.

[0007] The technical solution of the present invention is a method for constructing a soil factor microbial diversity database based on literature, comprising the following steps:

[0008] (1) Collect global soil factor-microorganism literature and determine the literature-related attributes, soil factor data, and microbial diversity data contained in the literature database;

[0009] (2) Extract the bibliographic information of each document and organize the bibliographic information data;

[0010] (3) Read the literature and appendices, extract the data on soil factors, pollution conditions, and microbial diversity and composition contained in the database, and organize the soil factor-microbial diversity data;

[0011] (4) Extract data from the figures and tables in the literature, use Origin software to extract specific values ​​from the data, and organize soil factor-microbial diversity data;

[0012] (5) sorting out the literature-related attributes, soil factors, and microbial diversity data extracted from each literature according to steps (2), (3), and (4) in Excel software, storing them in Access software, and forming a literature-based soil factor-microbial diversity database;

[0013] (6) The database can be updated, visualized and analyzed in Access software.

[0014] For database updates, you can use the Access crosstab query wizard to summarize the same date, or directly merge or import tables. You can also use the crowdsourcing platform users to input literature data, text statistics, and upload images to achieve data expansion.

[0015] For database visualization, you can use Access database to integrate data storage, data filtering and data visualization, and combine SQL, VBA and controls to quickly develop dynamic charts.

[0016] For data analysis, use SQL language in Access database to perform query analysis, distribution analysis or data map analysis.

[0017] Technical Effects

[0018] Compared with the prior art, the present invention constructs a soil microorganism database based on literature, which can provide users with convenient literature and data for query, especially for other research that may be conducted using artificial intelligence in the future. This database has the advantages of large amount, simplicity, global data coverage, multiple factor analysis and timeliness. In addition, it also uses the functions of the Access database itself to provide additional services such as update, visualization, data analysis, etc. to attract users to expand the database and apply the database. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a block diagram of an implementation of a method for constructing a soil factor-microorganism diversity database based on literature provided by the present invention. DETAILED DESCRIPTION

[0020] To make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation methods described herein are only used to explain the present invention and do not limit the protection scope of the present invention.

[0021] Based on the background technology, the present invention studies the key indicator parameters in the field of soil and soil microorganisms through investigation and inquiry, summarizes the basic information extraction indicators of the database including article publication information, research context and methods, soil microbial community indicators and soil indicators, stipulates data screening principles, and collects relevant papers from all over the world in the past eleven years (January 2010 to December 2021) as the data source, extracts literature information, literature content and attachment data and data in charts, records them in Excel, imports them into Access software to generate a database, and sets the literature serial number as the primary key field in Access to facilitate subsequent supplementary searches.

[0022] Subsequently, new papers can be found using the same search and screening methods, and the database can be updated in real time and summarized and supplemented using the Access cross-tab query wizard. Users of the crowdsourcing platform can also input literature information, literature content data, and upload images in the form of an online website, ultimately realizing a real-time, comprehensive, and scalable global soil factor-microorganism database knowledge acquisition platform.

[0023] See also Figure 1 The method for constructing a soil factor-microorganism diversity database based on literature provided in this embodiment comprises the following steps:

[0024] (1) Collect global soil factor-microbe literature and determine the literature-related attributes, soil factor data, and microbial diversity data contained in the literature database. We used "soil", "bacteri*", and "fung*" as keywords on the Web of Science website to search for a total of 422 publications in the past eleven years from January 2010 to December 2021. Based on existing papers and experience, the data we collected covered 27 indicators of soil microbial communities. In terms of microbial diversity, in addition to conventional diversity indicators including bacterial biomass (number of 16S genes), fungal biomass (number of ITS genes), bacterial diversity (Shannon index, Simpson index), and fungal diversity (Shannon index, Simpson index), we also collected the relative abundance of some common dominant species at the phylum level (such as Proteobacteria and Acidobacteria in bacteria, Ascomycota and Mortierella in fungi). These indicators can reflect the species composition of microorganisms and facilitate our analysis of the impact of pollutants on microbial species composition. In addition to microbial community indicators, the dataset also includes 21 soil factor indicators, including sampling time, soil location (latitude and longitude, altitude), average annual temperature, average annual precipitation, soil mechanical composition, sample depth, vegetation type (forest, shrub, grassland, tundra, wetland, desert, farmland), pH, soil total carbon, soil total organic matter, soil total nitrogen, and phosphorus. In addition, in order to study the impact of soil pollution on microbial diversity, we collected the types of soil pollution (heavy metal pollution, organic pollution, inorganic pollution, harmful microbial pollution, and radioactive pollution). These data were extracted from the literature text, attachments, and charts. When sorting the data, we screened according to certain conditions. The screening conditions for the data are: ① Only field studies are selected, excluding laboratory incubation studies; ② The data contains at least one microbial community indicator, including microbial biomass (16S gene and ITS gene number), microbial diversity (Shannon index, Simpson index), and the relative abundance of dominant species at the phylum level; ③ The data contains at least one type of microorganism (bacteria or fungi); ④ The microbial community structure is quantified by high-throughput sequencing technology; ⑤ The data contains at least one type of pollution

[0025] (2) Extract the bibliographic information of each document and organize the bibliographic information data; first extract the title and publication year of each document into an Excel spreadsheet, and arrange them according to a certain document number to facilitate subsequent import into Access for searching.

[0026] (3) Read the literature and the appendix, extract the soil factor and microbial diversity data contained in the database, screen them according to the screening principles, and organize the soil factor-microbial diversity data;

[0027] (4) Extract data from the figures and tables in the literature, use Origin software to extract specific values ​​from the figures and filter according to the screening principles, and organize soil factor-microbial diversity data. The image data processing software Origin is specifically used here. The specific operation steps are: open the image in the software, use the image digitization tool in the tool, manually select the coordinate axis, and enter the coordinate value in the figure. After the coordinate specification is completed, you can start manual point selection and manually select the points in the image where the values ​​should be extracted. After the point selection is completed, click the result, jump to the data, open the data list, and you can get the values ​​corresponding to the selected points. Record them in an Excel table and unify the format.

[0028] (5) After the data on the relevant attributes, soil factors, and microbial diversity extracted from each document according to steps (2), (3), and (4) are sorted out in Excel software, the data needs to be sorted out and supplemented. The specific operations include deleting and sorting out duplicate data in the Excel table, unifying all data formats under the same information category, such as longitude and latitude, all of which are unified as longitude E is positive and W is negative; latitude N is positive and S is negative; degrees, minutes, and seconds are converted to decimals (for example, longitude 120°18′ east is written as +120.3).

[0029] Different types of missing data were manually supplemented. The missing global soil properties (e.g., pH, total soil carbon, soil organic matter, total soil nitrogen, and total soil phosphorus) were supplemented in the database in the World Soil Information (ISRIC). The specific method is as follows: If the reference study did not report the latitude, longitude, or altitude of the experimental site, the approximate latitude and longitude were obtained by geocoding the location name in Google Earth 7.0 (free version). In the case that the reference study did not report the soil type, the values ​​were supplemented by the site geographic location Harmonized world soil database v1.2|FAOSOILS PORTAL|Food and Agriculture Organization of the United Nations website, Download viewer&data according to longitude and latitude.

[0030] If the reference study does not report pH and total soil nitrogen, search the URL: https: / / www.fao.org / soils-portal / data-hub / soil-classification / numerical-systems / chemical-properties / en / , roughly determine the location after opening the interface, refer to the legend, and take the average value in the interval. If the reference study does not report the soil organic carbon content, open the GSOCmap webpage, roughly determine the location based on the longitude and latitude, press the CTRL key and left click the mouse to select an area, and then take the mean value that appears on the left.

[0031] (6) After the unified format is completed, the corresponding literature information, literature content and attachment data, and icon data Excel table are imported into Access for storage to form a literature-based soil factor-microbial diversity database. The specific method is as follows: create a new Access database, select the Excel table option in the import data, select the Excel table, retain the first row of information containing the title, select the sorted literature number as the primary key to facilitate subsequent retrieval, and determine the name of the newly created database.

[0032] (7) The database can be updated, visualized and analyzed in Access software.

[0033] For database updates, you can use the Access crosstab query wizard to summarize the same date, or directly merge or import tables. You can also use the crowdsourcing platform users to input literature data, text statistics, and upload images to achieve data expansion.

[0034] For database visualization, use Access database to integrate data storage, data filtering and data visualization, and quickly develop dynamic charts by combining SQL, VBA and controls. Specifically, create a query first, insert a chart in the design mode of the form, set the data source to the query just created, create a static chart, and then set the chart ROWSOUCE = "SQL statement" in the VBA of the query button click event, and then replace part of the SQL statement with variables, so that you can create a dynamic chart.

[0035] For data analysis, use SQL language in Access database to perform query analysis, distribution analysis or data map analysis. Query analysis means creating a new "Query Design" in the "Create" tab, canceling the display table and entering the SQL dialog box, and using the query function SELECT to perform analysis. For distribution analysis, you must first create intervals in Access to mark them, create a new query, and use the Switch function in SQL to determine the distribution analysis within the interval. For data map analysis, first summarize the data in Access according to a certain requirement, paste the data into Excel, select and open Power MAP to generate a data map.

[0036] (8) Providing data foundation for the application of artificial intelligence such as machine learning

[0037] As mentioned in the "Background Technology" section, if we want to apply artificial intelligence methods such as machine learning to explore the relationship between environmental pollution, soil environmental factors and soil microbial diversity, we can screen and quantify the impact of important factors on soil microorganisms and predict microbial diversity and function at a global scale. We must establish a complete and comprehensive soil factor-microbial diversity database. This database can provide a data basis for the application of artificial intelligence such as machine learning. Here we briefly introduce a method for applying this database in machine learning.

[0038] We used three machine learning algorithms: random forest, artificial neural network, and support vector machine to train soil factor-microbial diversity regression models to explore the relationship between soil factors and microbial diversity. The data set was divided into five subsets according to different pollution types, and five regression models were established for each machine learning algorithm. The models were trained based on scikit-learn. The random forest model used 500 random decision trees and selected five random features at each node. The artificial neural network model is two-layer fully connected and can better learn nonlinear rules. The workflow is similar to that of the random forest. The support vector machine model selects radial basis function (RBF) as the kernel function and the regularization parameter is set to 1. For each model, "Adam" was selected as the optimizer to improve the convergence speed and parameter adjustment performance of the model. The ten-fold cross validation method was applied to estimate the average regression accuracy of the algorithm and calculate the 10-fold R between the predicted value and the observed value. 2 The average of the RMSE and root mean square error (RMSE) values ​​was used to measure the model performance.

Claims

1. A method for constructing a soil factor microbial diversity database based on literature, characterized in that: The steps include: 1) Research key indicator parameters in the field of soil and soil microorganisms; 2) Extraction indicators include article publication information, research context and methods, soil microbial community indicators, and basic information of the database of soil indicators; 3) Define data screening principles to collect data from massive amounts of historical soil microbial literature, store them in an Access database, and use the Access database to update, visualize and analyze the data; The details are as follows: (1) Determine the literature-related attributes, soil factor data, and microbial diversity data contained in the literature database, and collect soil factor / microbial literature from around the world; (2) Extract the bibliographic information of each document and organize the bibliographic information data; first extract the title and publication year of each document into an Excel table, and arrange them according to a certain document number to facilitate subsequent import into Access for searching; (3) Extract the soil factor and microbial diversity data contained in the database and screen them according to the screening principles to organize the soil factor-microbial diversity data; The data screening principles are as follows: ① Only field studies were selected, excluding laboratory incubation studies; ② The data contained at least one microbial community indicator, including: microbial biomass including at least 16S genes and ITS gene numbers; microbial diversity including Shannon index and Simpson index, and relative abundance of dominant species at the phylum level; ③ The data contained at least one type of microorganism such as bacteria or fungi; ④ The microbial community structure was quantified by high-throughput sequencing technology; ⑤ The data contained at least one type of pollution; (4) Extract data from figures and tables in the literature, use Origin software to extract data from the figures, and screen and organize soil factor / microbial diversity data according to the above screening principles; (5) Organize the literature-related attributes, soil factors, and microbial diversity data extracted from each literature according to steps (2), (3), and (4) in Excel, store them in Access, and form a literature-based soil factor-microbial diversity database; (6) After the unified format is completed, the corresponding literature information, literature content and attachment data, and icon data Excel table are imported into Access for storage to form a literature-based soil factor-microbial diversity database. The specific method is as follows: create a new Access database, select the Excel table option in the import data, select the Excel table, retain the first row of information containing the title, select the sorted literature number as the primary key to facilitate subsequent retrieval, and determine the name of the newly created database; (7) The database can be updated, visualized and analyzed in Access software.

2. The method for constructing a soil factor microbial diversity database based on literature according to claim 1, characterized in that: The step (5) specifically comprises: (1) Delete duplicate data in the Excel spreadsheet and unify the data format of all data under the same information category, such as longitude and latitude, all of which are unified as longitude E is positive and W is negative; latitude N is positive and S is negative; degrees, minutes and seconds are converted to decimals, for example, longitude 120°18′ east is written as +120.3; (2) Different types of missing data were manually supplemented. The missing global soil properties were supplemented in the database of the World Soil Information System (ISRIC) as follows: if the reference study did not report the latitude, longitude or altitude of the experimental site, the approximate latitude and longitude were obtained by geocoding the location name in Google Earth; (3) In cases where the reference study did not report the soil type, these values ​​were supplemented using the site geographic location Harmonizedworld soil database v1.2 | FAO SOILS PORTAL | Food and Agriculture Organization of the United Nations website, Download viewer & data, and search based on latitude and longitude; (4) If the reference study does not report pH and total soil nitrogen, search the URL: https: / / www.fao.org / soils-portal / data-hub / soil-classification / numerical-systems / chemical-properties / en / , roughly judge the location on the open interface, refer to the legend, and take the average number in the interval value; (5) If the reference study does not report the soil organic carbon content, open the GSOCmap webpage, roughly determine the location based on the longitude and latitude, press the CTRL key and left mouse button to select an area, and then take the mean value that appears on the left.

3. The method for constructing a soil factor microbial diversity database based on literature according to claim 1, characterized in that: In step (7), the database is updated, and the Access cross-tab query wizard is used to summarize the same date, or the tables are directly merged or imported. Subsequently, the crowdsourcing platform users can input literature data, text statistics, and upload images to achieve data expansion.

4. The method for constructing a soil factor microbial diversity database based on literature according to claim 1, characterized in that: In the step (7), the database visualization utilizes the Access database to integrate data storage, data screening and data visualization, and combines SQL, VBA and controls to quickly develop dynamic charts.

5. The method for constructing a soil factor microbial diversity database based on literature according to claim 1, characterized in that: The data analysis in step (7) uses SQL language to perform query analysis, distribution analysis or data map analysis in the Access database.

Citation Information

Patent Citations

  • Method for determining significant correlation between physicochemical property of rhizosphere soil of S-amino-acid-rich transgenic soybeans and diversity of microbial communities

    CN102994634A

  • A method for comprehensively analyzing polluted land parcel information based on literature

    CN109815315A