CN-121983279-A - Nut screening method for improving memory function based on key component target spot prediction and behavior verification

CN121983279ACN 121983279 ACN121983279 ACN 121983279ACN-121983279-A

Abstract

The invention relates to the technical field of food science, in particular to a nut screening method for improving memory function based on key component target spot prediction and behavior verification, which comprises the following steps of, based on behavior experiment record, analyzing the behavior change trend of the mice, comparing the trend with the GC-MS detection component scoring prediction performance, screening deviation performance, combining the target path with the signal path node category, and adjusting the target evaluation grouping to obtain a target grouping regulation result. According to the invention, by combining behavior trend quantification with target point function level weighting, multi-behavior data and component detection information are cooperatively processed, and by utilizing trend interval characteristics and target point evaluation parameter linkage screening, the causal relation between behavior change and functional components can be dynamically reflected, the structured quantitative expression of the action effect of nut components is realized, the systematicness and the judgment accuracy of the evaluation result are enhanced, the data result is stable and consistent, the screening period is shortened, and the scientific rationality and the application value of nut memory function screening are improved.

Inventors

GAO HAIYAN
SHEN CHAOYI
YIN MING
WU WEIJIE
CHEN HANGJUN
CHEN HUIZHI
LIU RUILING
FANG XIANGJUN
WANG GUANNAN
HAN YANCHAO
DENG YANGYONG

Assignees

浙江省农业科学院

Dates

Publication Date: 20260505
Application Date: 20260113

Claims (10)

1. The nut screening method for improving memory function based on key component target prediction and behavior verification is characterized by comprising the following steps: s1, analyzing daily behavior data of an experimental mouse in a platform crossing, target quadrant stay and latency experiment based on a behavior experiment record, judging a continuous variation interval of a behavior index, and obtaining a trend interval characteristic set; S2, comparing each behavior index trend with the gas chromatography-mass spectrum detection component score prediction performance based on the trend interval feature set, screening trend fragments with amplitude difference, and determining key behavior difference to obtain a deviation performance distribution item; S3, based on the deviation performance distribution item, the target path of the gas chromatography-mass spectrum detection data and the target prediction platform is called, and the node type of each component target in the KEGG signal path is judged to obtain a functional node sequence structure; s4, based on the functional node sequence structure, analyzing the functional effect of the difference nodes in the signal path, dividing the functional node sequence structure into growing, stabilizing and inhibiting groups according to the start, the relay and the tail end, optimizing the grouping classification, and obtaining functional class grouping data; And S5, screening functional target point evaluation items related to the deviation expression distribution items and the trend interval feature sets based on the functional category grouping data, analyzing the relation with the behavior change direction, and adjusting grouping weights to obtain target point grouping regulation and control results.
2. The method for screening nuts with improved memory based on key component target prediction and behavior verification according to claim 1, wherein the trend interval feature set comprises dominant trend type, interval variation amplitude and trend persistence, the deviation expression distribution item comprises variation direction distribution, difference amplitude distribution and key deviation feature, the functional node sequence structure comprises node functional labels, hierarchical ordering numbers and node association relations, the functional category grouping data comprises growth category information, stability category information and inhibition category information, and the target point grouping regulation result comprises evaluation weight distribution, trend adjustment factors and regulation classification identification.
3. The method for screening nuts with improved memory function based on key component target prediction and behavior verification according to claim 1, wherein the step of trend interval feature set is specifically as follows: S111, based on behavior experiment records, daily data of experiment mice in platform crossing, target quadrant stay and latency experiment are sorted, data performances of all behavior indexes of all mice in continuous dates are compared, and variation trend of all the indexes is calculated by comparing variation directions of the different daily index performances item by item, so that a behavior trend sequence set is obtained; s112, classifying trend directions of various indexes of each mouse based on the behavior trend sequence set, respectively identifying time periods with continuously enhanced, continuously weakened or tending to be stable, and classifying trends of the identified time periods to obtain a trend segmentation information set; s113, based on the trend segmentation information set, comparing trend identifications of all behavior indexes in all time segments, screening trend types with more occurrence times in each index, and sorting dominant trend directions and time segment information of each index to obtain a trend interval feature set.
4. The method for screening nuts with improved memory function based on key ingredient target prediction and behavior verification according to claim 1, wherein the step of deviating from the performance distribution item specifically comprises the following steps: S211, based on the trend interval feature set, comparing the change direction of each behavior index with the prediction direction of the component score corresponding to the gas chromatography-mass spectrum detection data, analyzing the consistency between each pair of direction labels, and recording consistent and inconsistent direction pairing information according to the behavior index to obtain a trend consistency sequence; S212, based on the trend consistency sequence, screening behavior indexes with inconsistent directions, analyzing time sequence performances of the indexes in original behavior data, judging performance changes in each time interval, and inducing various behavior fluctuations through marking jump fragments to obtain a fluctuation fragment information group; S213, judging the expression structure of each behavior index based on the fluctuation segment information set, analyzing the structural continuity and fluctuation amplitude in the jump segment, and identifying the segment with the expression change and the prominent sequence characteristic to obtain the deviation expression distribution item.
5. The method for screening nuts with improved memory function based on key component target prediction and behavior verification according to claim 1, wherein the step of the functional node sequence structure is specifically as follows: s311, analyzing the nut active ingredients based on the deviation expression distribution item, retrieving the gas chromatography-mass spectrum detection result and target point data of the target point prediction platform, judging the action target point numbers corresponding to the ingredients, and obtaining ingredient target point index data; S312, comparing each target point with signal path data in a KEGG database based on the component target point index data, analyzing the initial, relay and end positions of each target point in the signal path structure, and dividing functional layers according to the arrangement sequence of path nodes to obtain a target point layer label set; S313, sequentially summarizing targets corresponding to each nut active ingredient based on the target level label set, classifying and carding functional levels, and sorting structural sequences of the targets in the signal path to obtain a functional node sequence structure.
6. The method for screening nuts with improved memory function based on key component target prediction and behavior verification according to claim 1, wherein the step of grouping data of functional categories specifically comprises: S411, analyzing the arrangement and the number of nodes of each signal path based on the functional node sequence structure, judging the initial, relay or end functions of each node on the path, classifying the node function sequence, and obtaining a node function label group; s412, based on the node function label group, comparing the distribution of the nodes of each role type in the path, adjusting target evaluation groups according to the role functions, dividing the initial nodes into growing categories, dividing the relay nodes into stable categories, and dividing the end nodes into inhibition categories to obtain target evaluation classification data; s413, based on the target evaluation classification data, adjusting a target evaluation item structure, merging the grouping label and the function category, and executing data grouping according to the node function attribute to obtain function category grouping data.
7. The method for screening nuts with improved memory function based on key component target prediction and behavior verification according to claim 1, wherein the step of grouping the target regulation results is specifically as follows: S511, screening target point evaluation items related to the deviation performance distribution items and the trend interval feature sets based on the function category grouping data, analyzing the direction corresponding relation between each target point evaluation item and the behavior change trend, and counting the performance categories of each item in the difference index to obtain a target point attribution distribution set; S512, based on the target attribution distribution set, comparing the behavior direction distribution of target evaluation items of each category, analyzing the direction distribution situation of the target evaluation items in each function category, and carrying out weight adjustment by combining with a grouping label to obtain a grouping weight parameter set; S513, based on the grouping weight parameter set, adjusting weight distribution of each grouping, optimizing the ordering structure of target evaluation items under each function category, and re-summarizing the influence degree of each grouping target evaluation item to obtain a target grouping regulation result.
8. The nut screening method for improving memory function based on key component target prediction and behavior verification according to claim 1, wherein the gas chromatography-mass spectrometry quantitatively analyzes the content of key components in different nut samples, the key components are obtained by combining network topology parameter analysis to construct a compound network of nut component targets, and the compound network of nut component targets is subjected to network topology screening to obtain screening criteria including the number of acting targets, connection density and association degree with memory function targets.
9. The method for screening nuts with improved memory function based on key ingredient target prediction and behavioral verification according to claim 8, wherein said key ingredients include linoleic acid, α -linolenic acid, arachidonic acid, heptadecenoic acid, palmitic acid, arachidonic acid and decanoic acid.
10. The method for screening nuts with improved memory function based on key component target prediction and behavior verification according to claim 1, wherein the platform crossing refers to the behavior of a mouse crossing an area where a stealth platform is located in a Moris water maze experiment, and the score prediction performance refers to a prediction performance range or interval calculated by using the content of key components detected by GC-MS and a corresponding target MPSI model.

Description

Nut screening method for improving memory function based on key component target spot prediction and behavior verification Technical Field The invention relates to the technical field of food science, in particular to a nut screening method for improving memory function based on key component target spot prediction and behavior verification. Background The food science mainly researches the composition, structure and property of food and the laws of physical, chemical and biological changes in the processing, storage and consumption processes, covers a plurality of matters such as screening and evaluation of functional ingredients of food raw materials, research on nutrition and health mechanisms, optimization of food processing technology, detection and control of food safety, development of functional food and the like, emphasizes that ingredients with health regulation function in natural food materials are identified and evaluated through a scientific method so as to guide accurate nutrition intervention and diet structure optimization. The traditional nut screening method is a method for judging whether the nut food in the dietary components has memory improving potential or not by taking the nut food in the dietary components as objects and adopting an empirical judgment or single animal experiment mode and evaluating the influence on the cognitive function of an organism after ingestion. In the prior art, animal experiments and single behavior index analysis are adopted to judge the memory improvement effect of nuts, comprehensive excavation on trend changes of multiple groups of experimental data is lacking, actual interaction between key components and multidimensional behavior can not be considered in an operation flow, so that component screening randomness is large, data results are obviously influenced by experimental contingencies, the relation between nut components and cognitive functions is difficult to reveal from a signal path layer, evaluation process scientificity and repeatability are insufficient, and problems of long flow and inaccurate screening conclusion exist in functional food development links. Disclosure of Invention In order to solve the technical problems in the prior art, the embodiment of the invention provides a nut screening method for improving memory function based on key component target spot prediction and behavior verification. The technical scheme is as follows: in one aspect, a method for screening nuts with improved memory function based on key component target prediction and behavior verification is provided, comprising the following steps: s1, analyzing daily behavior data of an experimental mouse in a platform crossing, target quadrant stay and latency experiment based on a behavior experiment record, judging a continuous variation interval of a behavior index, and obtaining a trend interval characteristic set; S2, comparing each behavior index trend with the gas chromatography-mass spectrum detection component score prediction performance based on the trend interval feature set, screening trend fragments with amplitude difference, and determining key behavior difference to obtain a deviation performance distribution item; S3, based on the deviation performance distribution item, the target path of the gas chromatography-mass spectrum detection data and the target prediction platform is called, and the node type of each component target in the KEGG signal path is judged to obtain a functional node sequence structure; s4, based on the functional node sequence structure, analyzing the functional effect of the difference nodes in the signal path, dividing the functional node sequence structure into growing, stabilizing and inhibiting groups according to the start, the relay and the tail end, optimizing the grouping classification, and obtaining functional class grouping data; And S5, screening functional target point evaluation items related to the deviation expression distribution items and the trend interval feature sets based on the functional category grouping data, analyzing the relation with the behavior change direction, and adjusting grouping weights to obtain target point grouping regulation and control results. On the other hand, the trend interval feature set comprises a dominant trend type, interval variation amplitude and trend persistence, the deviation expression distribution item comprises variation direction distribution, difference amplitude distribution and key deviation features, the function node sequence structure comprises node function labels, hierarchical ordering numbers and node association relations, the function class group data comprises growth class information, stable class information and inhibition class information, and the target point group regulation and control result comprises evaluation weight distribution, trend adjustment factors and regulation and control classification identifiers. On the other hand, the step of the tr