Edible mushroom breeding method

Abstract

The invention provides an edible mushroom breeding method, which belongs to the technical field of edible mushroom cultivation, and comprises the following steps: carrying out physical mutagenesis treatment on original strains of edible mushrooms, carrying out multiple rounds of low-intensity accumulative mutagenesis to obtain a mutant strain library, and then applying selection pressure associated with target characters to the mutant strain library to obtain a mutant strain library; the method comprises the following steps: collecting a cultured bacterial colony image, extracting quantitative characteristics as an agent phenotype, analyzing the agent phenotype by utilizing a machine learning model, screening out a primary candidate strain, then carrying out single spore isolation and karyotype identification on the primary candidate strain to obtain homonyons, and carrying out controlled pairing or backcross on the homonyons with target characters to obtain advanced candidate strains; and finally, performing multi-generation cultivation under the conditions of simulating actual production and including environmental disturbance, and screening out a final strain by taking the yield and the character stability as indexes. The method solves the problems of single genetic background of mutants, low screening accuracy and efficiency, insufficient genetic stability of strains and poor industrial adaptability in existing edible mushroom breeding.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of edible mushroom cultivation, in particular to a breeding method of edible mushroom. BACKGROUND

[0002] With the continuous expansion of the industrialization scale of edible mushrooms, the market has higher requirements for the yield, quality, stress resistance and genetic stability of strains. Edible mushroom breeding technology is rapidly developing towards directionality, high efficiency and stability. Traditional breeding methods are gradually integrated with modern biological technology and artificial intelligence technology. Physical mutagenesis, with the advantages of simple operation, wide mutation spectrum and no chemical residues, has become one of the core means for obtaining mutants in edible mushroom breeding. The introduction of machine learning and molecular identification technologies also makes it possible to accurately control the breeding process and promote the transformation of breeding mode from "random mutagenesis-blind screening" to "directional mutagenesis-accurate screening".

[0003] At present, a multi-technology system based on physical / chemical mutagenesis combined with hybrid breeding and molecular marker-assisted breeding has been formed in the field of edible mushroom breeding. Among them, physical mutagenesis methods such as normal temperature and pressure plasma, low-energy ion beam and ultraviolet light have been widely used in the mutation treatment of starting strains; single spore isolation, lock-like joint observation and molecular marker detection technologies are also gradually used for strain purification and karyotype identification; some studies attempt to extract phenotypic characteristics such as colony diameter and mycelium density to assist in screening candidate strains with target traits, which to some extent provides technical support for improving breeding efficiency.

[0004] However, the existing technology still has significant limitations: first, physical mutagenesis mostly uses single high-intensity dose treatment mode, which can increase the mutation rate in a short period of time, but often accompanied by high mortality rate and easy to lead to homogenization of the genetic background of mutants, and there is a lack of mature solution for balancing the dose of mutagen and the survival of strains and mutation diversity; second, the quantitative analysis of related characteristics in the phenotypic screening link is not deep enough, and mostly relies on manual observation or simple parameter measurement, which has strong subjectivity and insufficient data precision; third, the verification link of the transmission stability of strains after pairing and hybridization is weak, and the number of subculture rounds in some breeding processes is limited, which makes it difficult to fully guarantee the genetic persistence of target traits; fourth, the cultivation screening stage is mostly carried out under controllable ideal environmental conditions, and the environmental fluctuations that may occur in actual production scenarios are not considered, which leads to the problem that excellent strains obtained through laboratory screening often face insufficient adaptability and large yield fluctuations in industrialization application, affecting the actual transformation effect of breeding results. SUMMARY

[0005] To overcome the shortcomings of existing technologies, the purpose of this invention is to provide a breeding method for edible fungi that can significantly broaden the genetic diversity of mutants, improve the accuracy and efficiency of target trait screening, ensure the genetic stability of target traits, enhance the adaptability of strains to industrial environments, and thus significantly improve the conversion efficiency of breeding results to large-scale production, ultimately cultivating excellent edible fungi strains with superior comprehensive agronomic traits that meet the requirements of industrial mass production.

[0006] To achieve the above objectives, the present invention provides the following solution: A method for breeding edible fungi, comprising the following steps: S1. Physical mutagenesis was performed on the starting strains of edible fungi, and based on the dose-survival rate relationship determined in the preliminary experiment, multiple rounds of low-intensity cumulative mutagenesis were carried out within the sublethal dose window to obtain a mutant strain library with rich genetic background. S2. Apply selection pressure associated with the target trait to the obtained mutant strain library, extract quantitative features based on the colony images collected during the culture process as surrogate phenotypes, and use machine learning models to analyze the surrogate phenotypes and screen out primary candidate strains. S3. The selected primary candidate strains are isolated by single spore isolation and karyotype identification to obtain homokaryosomes. Homokaryosomes with the target traits are then subjected to controlled pairing or backcrossing to obtain advanced candidate strains with stable genetic backgrounds. S4. Based on the obtained advanced candidate strains, multiple generations of cultivation are carried out under conditions that simulate actual production and include environmental disturbances. The final strain that meets the requirements for industrialization is selected by using yield and trait stability as indicators.

[0007] Preferably, in S1, the physical mutagenesis treatment is selected from one or more of ambient temperature and pressure plasma, low-energy ion beam, or closed ultraviolet mutagenesis.

[0008] Preferably, in S1, the dose-survival relationship determined based on preliminary experiments is achieved in the following manner: Set up gradient mutagenesis doses to treat the starting strain, plot the dose-survival curve, and use the dose range corresponding to a survival rate of 20%-40% as the sublethal dose window.

[0009] Preferably, in S2, the surrogate phenotype includes one or more of the following: daily growth rate of colony diameter, colony edge regularity, and hyphal density.

[0010] Preferably, in S2, the machine learning model is a regression or classification model constructed using a support vector machine or random forest algorithm.

[0011] Preferably, in S3, the single spore isolation is achieved by plate dilution or micromanipulation, and the karyotype identification is achieved by microscopic observation of clamp connections or by PCR amplification detection using molecular markers targeting mating sites.

[0012] Preferably, in S3, after the controlled pairing or backcross, the resulting strain is subjected to at least three rounds of subculture to verify the passage stability of the target trait.

[0013] Preferably, in S4, the environmental disturbance includes diurnal temperature fluctuations of 1-3°C.

[0014] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects: (1) The edible fungi breeding method provided by the present invention uses multiple rounds of low-intensity cumulative physical mutagenesis within a sublethal dose window. Compared with the traditional single high-intensity mutagenesis, it avoids the limitation of high lethality on the size of the strain library and greatly expands the genetic diversity of mutants, laying a rich genetic foundation for the discovery of high-quality target traits. At the same time, it uses quantitative proxy phenotypes such as daily growth of colony diameter and mycelial density combined with support vector machine or random forest machine learning models to carry out screening, replacing the traditional manual subjective observation method, significantly improving the accuracy and efficiency of target trait screening, and realizing the rapid locking of primary candidate strains from a large number of mutant strains.

[0015] (2) This invention relies on the plate dilution method or micromanipulation method for single-spore isolation, clamp joint observation or karyotype identification by molecular markers of mating type sites to accurately screen out homozygous congeners. Then, by controlling pairing / backcrossing and subculture for no less than three rounds, strains with mixed genetic backgrounds can be effectively eliminated, strongly ensuring the intergenerational genetic stability of the target trait. This avoids the trait degradation problem caused by impure strain karyotype and insufficient subculture verification in traditional breeding from the root, ensuring that excellent traits can be stably inherited.

[0016] (3) This invention conducts simulated production of advanced candidate strains through multiple generations of cultivation with environmental disturbances such as diurnal temperature fluctuations of 1-3℃, and uses yield and trait stability as the core indicators for final screening. Compared with the traditional cultivation verification under ideal conditions, it is closer to the actual industrial production scenario and can accurately screen out the final strains with strong adaptability to complex production environments, which greatly improves the conversion efficiency of breeding results to large-scale production. The strains finally cultivated not only have outstanding target traits, but also meet the core requirements of industrial mass production, providing key strain support for the high-quality and large-scale development of the edible fungi industry. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating a breeding method for edible fungi according to the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0021] Example 1 like Figure 1 As shown in the figure, this embodiment provides a method for breeding edible fungi, including the following steps: S1. Physical mutagenesis was performed on the starting strains of edible fungi, and based on the dose-survival rate relationship determined in the preliminary experiment, multiple rounds of low-intensity cumulative mutagenesis were carried out within the sublethal dose window to obtain a mutant strain library with rich genetic background.

[0022] Furthermore, step S1 specifically includes: The starting strain was activated to the logarithmic growth phase to prepare a concentration of 1×10⁻⁶. 6 ~1×10 7 A spore suspension of 100 spores / mL is used to ensure the uniformity of the mutagenesis treatment and support the stability of subsequent mutation effects. Physical mutagenesis treatment can be performed using one or more of the following: ambient temperature and pressure plasma, low-energy ion beam, or closed ultraviolet mutagenesis. The principle is to act on the genetic material of the strain through physical factors, inducing mutations such as DNA strand breaks, base substitutions, or deletions. Different physical mutagenesis methods have different mutagenic sites; combining them can further broaden the mutation spectrum. Simultaneously, a control system is constructed by setting 5-7 gradient mutagenesis doses. For ambient temperature and pressure plasma, dose gradients of 10s, 20s, 30s, 40s, 50s, and 60s can be set; for low-energy ion beam, a dose gradient of 5 × 10⁻⁶ can be used. 13 ions / cm 2 1×10 14 ions / cm2 1.5×10 14 ions / cm 2 2×10 14 ions / cm 2 2.5×10 14 ions / cm 2 Dose gradients were used, with three replicates for each dose group and a blank control group.

[0023] Spore suspensions from each dosage group were evenly spread on PDA agar plates and incubated in the dark at 25°C for 7 days. Colony counts were then recorded, and the survival rate of the strains at different dosages was calculated using the formula (treatment group colony count / control group colony count × 100%). A dose-survival rate curve was plotted, and the dosage range of 20%-40% survival rate was selected as the sublethal dose window. Precise determination of this range is crucial for balancing mutation rate and strain survival rate. Below 20% survival rate, high-intensity mutagenesis can easily lead to excessive genetic damage to the strains, resulting in very few effective mutants surviving. Above 40% survival rate, the mutagenesis intensity is insufficient, making it difficult to generate enough target trait mutation sites.

[0024] The median of the sublethal dose window was selected as the single mutagenesis dose, and the starting strain was subjected to 3-5 rounds of repeated mutagenesis. After each round of mutagenesis, the surviving strains were collected to prepare spore suspensions again. The principle is that low-intensity mutagenesis only induces a small amount of genetic material variation each time, and the strain's repair mechanism can avoid lethal damage. Multiple rounds of accumulation can superimpose mutations at different sites, thereby enriching genetic diversity. Compared with traditional single high-intensity mutagenesis, it can significantly increase the number of effective mutants and provide a sufficient genetic basis for subsequent screening of target traits.

[0025] S2. Apply selection pressure associated with the target trait to the obtained mutant strain library, extract quantitative features based on colony images collected during the culture process as surrogate phenotypes, and use machine learning models to analyze the surrogate phenotypes and screen out primary candidate strains.

[0026] Furthermore, step S2 specifically includes: Selective culture media are configured according to specific breeding objectives to apply targeted selection pressure. If the breeding target is high-yielding mycelium, the carbon-to-nitrogen ratio of the culture medium is adjusted to 20:1–25:1, while the conventional medium has a carbon-to-nitrogen ratio of 15:1. This increases the carbon source ratio to meet the energy requirements of high-yielding strains. If the target is low-temperature tolerance, the culture temperature is controlled at 15°C, while the conventional culture temperature is 25°C. This low-temperature stress enriches strains with cold-resistant metabolic pathways. The principle is that selective pressure inhibits the growth of non-target strains, selectively retaining strains with the potential for the target trait, and reducing the workload of subsequent screening.

[0027] High-resolution imaging equipment with a resolution of ≥10 megapixels was used to collect colony images daily starting from day 3 post-inoculation. Surrogate phenotypes were extracted using image analysis software, allowing selection of one or more of the following: daily colony diameter growth, colony edge regularity, or hyphal density. Daily colony diameter growth was obtained by measuring colony diameter daily and calculating the difference between adjacent days. Colony edge regularity was calculated using an edge detection algorithm to determine the contour fit (values ​​from 0 to 1). Hyphae density was calculated using grayscale analysis to determine the percentage of hyphae pixels per unit area (values ​​from 0 to 100%). These surrogate phenotypes showed significant correlations with the target traits; for example, hyphal density was positively correlated with high yield, and edge regularity was correlated with strain growth stability. This transformed the qualitative morphology obtained through traditional subjective observation into precise and quantifiable data, avoiding errors from human judgment. Using "possession of the target trait" as a label, the extracted surrogate phenotypic data are divided into training and test sets in a 7:3 ratio. Support Vector Machines (SVMs) or Random Forest algorithms are used to construct regression or classification models. Random Forest models can be configured with 100-200 decision trees and a maximum depth of 5-8 layers. The model establishes discrimination rules by learning the correspondence between surrogate phenotypic patterns and target traits in the training set. After the model training is complete and the test set accuracy is ≥90%, the surrogate phenotypic data of the strains to be screened are input into the model, enabling rapid and accurate sorting of massive numbers of mutant strains. Compared to manual screening, which requires observation and measurement of each strain, the model can process data in batches, and the discrimination based on quantitative indicators avoids subjective bias, significantly improving screening efficiency. At the same time, the test set accuracy of over 90% ensures the reliability of the screening results, effectively reducing the missed screening rate and redundant operations.

[0028] S3. The selected primary candidate strains are isolated by single spore isolation and karyotype identification to obtain homokaryotes. Homokaryotes with the target traits are then subjected to controlled pairing or backcrossing to obtain advanced candidate strains with stable genetic backgrounds.

[0029] Furthermore, step S3 specifically includes: Single spore isolation was performed using either the plate dilution method or the micromanipulation method. The plate dilution method required serially diluting the spore suspension of the primary candidate strain to 10-1. -6 ~10 -7The concentration was g / mL. 100 μL of the diluted solution was spread onto a PDA plate and incubated in the dark at 25°C for 2 days. Single colonies were then picked under a microscope. If the spore yield of the strain was low, single spores were directly picked and cultured using micromanipulation. The principle is that the genetic material of the strain formed by the germination of a single spore is homozygous, which can avoid the segregation of traits in subsequent generations of heterozygous strains. Karyotype identification was performed using a dual verification method. On the one hand, a single colony mycelial slide was prepared, and clamp connections were observed under an optical microscope. Homokaryotic cells do not have clamp connections, while heterokaryotic cells form clamp connections due to hyphal fusion to maintain the exchange of genetic material. This is the classic morphological basis for fungal karyotype identification. On the other hand, specific primers were designed for mating type loci of edible fungi (such as the A and B mating type loci of shiitake mushrooms) to amplify mycelial DNA by PCR. Homokaryotic cells contain only one mating type genotype, and the amplification product shows only one band. Heterokaryotic cells contain two mating type genotypes, and the amplification product shows two bands. Dual verification can accurately distinguish kernel types, avoid misjudgment by a single method, and retain only pure contract kernels for subsequent steps.

[0030] In addition, different mating type homokaryotes possessing the target trait were inoculated into the same PDA plate at a 1:1 ratio for controlled pairing, or backcrossed with the starting strain. The principle is that homokaryotes fuse to form heterokaryotes, which can integrate the superior traits of both parents, while backcrossing can retain the target trait while stabilizing the genetic background. After mycelial fusion, the resulting strain was subjected to at least three rounds of subculture, with each round lasting 7 days, and the target trait was measured in each round. The genetic stability of the target trait in edible fungi needs to be verified through multiple rounds of subculture. Three rounds of subculture can cover the common genetic fluctuation cycle during strain propagation. If the coefficient of variation of the target trait is <5% after three rounds of subculture, it indicates that the trait is genetically controlled and stable, and is not prone to phenotypic drift due to propagation. This threshold is based on the industry-standard verification of genetic stability in edible fungi breeding. Strains with a genetic background below this threshold have a very low risk of trait degradation in subsequent cultivation and are therefore identified as advanced candidate strains with stable genetic backgrounds.

[0031] S4. Based on the obtained advanced candidate strains, multiple generations of cultivation are carried out under conditions that simulate actual production and include environmental disturbances. The final strain that meets the requirements for industrialization is selected by using yield and trait stability as indicators.

[0032] Furthermore, step S4 specifically includes: A cultivation substrate consistent with actual industrial production is used. For example, the shiitake mushroom cultivation substrate is prepared with a ratio of sawdust:wheat bran:corn cob = 70:20:10, and the moisture content is controlled at 60%~65%. Advanced candidate strains are inoculated into standard cultivation bags to ensure consistency between cultivation conditions and industrial production, thus guaranteeing the practicality of the screening results. The principle is that the nutrient composition and physical structure of the cultivation substrate directly affect the growth and yield of the strains. Conditions consistent with industry can avoid screening deviations caused by differences between laboratory and actual production environments. A cultivation environment that closely matches the actual production scenario is constructed, controlling the humidity at 85%~90%, the light cycle at 12h light / 12h dark, and setting a core diurnal temperature fluctuation of 1-3℃, such as 25℃ during the day and 22℃ at night. In actual industrial production, diurnal temperature variations are difficult to avoid. These fluctuations affect the metabolic enzyme activity and growth rhythm of the strains. Simulating these conditions can screen for strains with strong metabolic regulation capabilities and resistance to environmental fluctuations, avoiding the industrial yield fluctuations caused by poor environmental adaptability of strains screened under traditional ideal constant environments.

[0033] Advanced candidate strains were continuously cultivated for 3-5 generations. The intergenerational variation coefficient of single-bag fresh mushroom yield and target traits was measured for each generation. Multiple generations of cultivation reflect the long-term genetic stability of the strains and avoid the randomness of single-generation cultivation. The improvement in average single-bag yield compared to the starting strain was determined by referencing the conventional effective improvement range of existing edible fungi mutation breeding, combined with the targeted enrichment effect of multiple rounds of screening in this embodiment, to set a reasonable screening threshold. The intergenerational variation coefficient of the target trait was <3% because industrial production requires phenotypic consistency across multiple generations of cultivation. This threshold is lower than the acceptable trait fluctuation range in conventional production, typically 5%, ensuring stable yield and quality in large-scale production. Combining these two indicators, the final strain, possessing both outstanding target traits and meeting the needs of industrial mass production, was selected, significantly improving the efficiency of converting breeding results into large-scale production.

[0034] The above content will be further explained below through specific implementation methods. The described embodiments are only some embodiments of the present invention.

[0035] Example 2 This embodiment uses the conventional shiitake mushroom production strain L808 as the starting strain, with an average yield of 1.20 kg per bag and a generational variation coefficient of 6.8%. The experiment was set up with a control group of the prior art, with 3 replicates in each group and 100 standard cultivation bags (17cm×33cm, 1.8kg / bag) per replicate. The method provided in Example 1 was used in the embodiment group, while the conventional ultraviolet mutagenesis breeding method was used in the prior art control group.

[0036] Specifically, the experimental steps of this embodiment group include: First, multiple rounds of low-intensity cumulative physical mutagenesis were performed within a sublethal dose window to activate the L808 strain of *Lentinula edodes* to the logarithmic growth phase, preparing a solution with a concentration of 1.5 × 10⁻⁶. 6 Spore suspensions of 1000 cells / mL were used, and room-temperature, ambient-pressure plasma was selected as the physical mutagenesis method. Six dose gradients were set at 10s, 20s, 30s, 40s, 50s, and 60s, with three replicates for each dose group and a blank control group. The spore suspensions of each dose group were evenly spread on PDA medium plates and incubated in the dark at 25℃ for 7 days. After that, the colony count was counted, the survival rate was calculated, and a dose-survival rate curve was plotted. The 30s corresponding to a survival rate of 30% was determined as the median value of the sublethal dose window. The starting strain was subjected to four rounds of cumulative mutagenesis at this dose. After each round of mutagenesis, the surviving strains were collected and spore suspensions were prepared again, finally obtaining a strain library containing 320 effective mutants.

[0037] Next, machine learning-assisted quantitative surrogate phenotypic screening was conducted. With "high yield" as the breeding objective, a selective medium with a carbon-to-nitrogen ratio of 25:1 was prepared, while the conventional medium had a carbon-to-nitrogen ratio of 15:1. Strains from the mutant strain library were inoculated into this medium and cultured in the dark at 25°C. Starting from the 3rd day after inoculation, colony images were collected daily using a 12-megapixel imaging device. Two quantitative surrogate phenotypic traits, namely the daily growth rate of colony diameter and mycelial density, were extracted using image analysis software. With "single bag yield ≥ 1.3 kg" as the target trait label, the extracted phenotypic data were divided into training and test sets in a 7:3 ratio. A random forest classification model with 150 decision trees and a maximum depth of 6 layers was constructed. After the model was trained, the accuracy of the test set was 82%. The surrogate phenotypic data of all mutant strains were input into the model, and 76 primary candidate strains were selected.

[0038] Subsequently, single-spore isolation, karyotype identification, and genetic stability verification were carried out. The spore suspensions of the primary candidate strains were serially diluted to 10⁻⁶ using the plate dilution method. -7 100 μL of the mycotoxin was plated on a PDA plate and incubated in the dark at 25°C for 2 days. Single colonies were then picked and homozygous congeners were screened by double karyotype identification. No clamping junctions were observed under an optical microscope. Specific primers were designed for PCR amplification targeting mating sites A and B of shiitake mushrooms. Only one band appeared in the amplification product, and 31 homozygous congeners were finally obtained. Homozygous congeners of different mating types with high yield potential were inoculated into the same PDA plate at a 1:1 ratio for controlled pairing. After mycelial fusion, three rounds of subculture were performed (7 days per round). Mycelial growth rate and yield-related enzyme activities were measured in each round. After three rounds of subculture, 24 strains with a coefficient of variation of target traits <5% were identified as advanced candidate strains.

[0039] Finally, a multi-generational cultivation screening was conducted in a simulated industrial environment. Using a conventional industrial cultivation substrate (sawdust: wheat bran: corn cob = 70:20:10, moisture content 62%), 24 advanced candidate strains were inoculated into standard cultivation bags to create a simulated production environment. The humidity was controlled at 88%, the light cycle was 12h light / 12h dark, and the diurnal temperature fluctuation was set at 2℃ (e.g., 25℃ during the day and 23℃ at night). The advanced candidate strains were continuously cultivated for 4 generations. The yield of fresh mushrooms per bag and the intergenerational variation coefficient of the target trait were measured for each generation. Finally, one high-yielding strain, L808-1, was selected, which met the requirements. Its average yield per bag over 4 generations was 1.45kg, and the intergenerational variation coefficient of the target trait was 4.2%.

[0040] The existing control group uses the traditional ultraviolet mutagenesis breeding method, specifically as follows: A spore suspension of the L808 shiitake mushroom strain was prepared and subjected to a single high-intensity mutagenesis irradiation with 60s ultraviolet light (8% survival rate), yielding 152 mutant strains. These mutants were inoculated into conventional PDA medium and cultured in the dark at 25℃. Colony morphology and mycelial growth were observed manually, and 45 candidate strains were selected based on experience. The selected strains were directly paired and hybridized, and after only one round of subculture, they were inoculated into cultivation bags. One generation of cultivation was carried out under undisturbed conditions of constant 25℃, 88% humidity, and a 12h light / 12h dark light cycle, resulting in the L808-control strain, with an average yield of 1.35 kg per bag and a generational variation coefficient of 8.7%.

[0041] The experimental data of this embodiment group and the prior art control group were summarized, and the results are shown in Table 1.

[0042] Table 1 Comparison of Experimental Data Results

[0043] It should be noted that the industry standard data is based on the following: 1. The average yield per bag is 1.20 kg, which is calculated by referring to the conventional production yield range of cultivated strains in "Exploration of Germplasm Resources for Facility-Based Shiitake Mushrooms" (6.38-566.4 g / small-sized cultivation bottle) and combining it with the industry-standard yield of a standard cultivation bag with a filling capacity of 1.8 kg; 2. The intergenerational variation coefficient of the trait is 6.8%, which meets the variation coefficient requirement of conventional stable strains in "GBT21125-2007 Technical Specifications for the Breeding of Edible Fungi" (≤10%), and is also derived by referring to the growth period variation coefficient of facility-based cultivated shiitake mushroom strains (5.63).

[0044] As shown in Table 1, the effective number of mutant strains in this embodiment was 2.1 times that of the control group through multiple rounds of low-intensity cumulative mutagenesis, thus reasonably broadening genetic diversity. Using quantitative surrogate phenotypes and machine learning screening, the accuracy rate was improved by 17 percentage points compared to the control group, effectively reducing the subjective error of manual screening. Through single-spore isolation, karyotype identification, and three rounds of subculture, the trait stability rate was improved by 36 percentage points compared to the control group, significantly reducing the risk of trait degradation during generations. Multi-generation cultivation screening simulating environmental disturbances made the strains more suitable for industrial production, with a yield increase of 20.8% compared to the starting strain, higher than the 12.5% ​​increase in the control group. Furthermore, the intergenerational variation coefficient of the trait was lower than that of the control group and the industry standard, fully demonstrating the significant technical advantages of this embodiment in high-yield strain breeding.

[0045] Example 3 This embodiment uses the common strain F47 of *Flammulina velutipes* as the starting strain. Its average yield per bag at 15℃ is 0.70 kg, and the survival rate at low temperature is 65%. The experiment included this embodiment group and a control group using existing technology, with three replicates in each group. Each replicate contained 100 standard cultivation bags (15cm × 30cm, 1.2kg / bag). This embodiment group used the method provided in Example 1, while the control group used traditional chemical mutagenesis breeding methods.

[0046] Specifically, the experimental steps of this embodiment group include: First, multiple rounds of low-intensity cumulative physical mutagenesis were performed within a sublethal dose window to activate the *Flammulina velutipes* F47 strain to the logarithmic growth phase, preparing a concentration of 2.0 × 10⁻⁶. 6 A spore suspension of 10 spores / mL was used, and a low-energy ion beam was selected as the physical mutagenesis method, with a setting of 5 × 10⁶ ions / mL. 13 ions / cm 2 1×10 14 ions / cm 2 1.5×10 14 ions / cm 2 2×10 14 ions / cm 2 2.5×10 14 ions / cm 2 Five dose gradients were used, with three replicates for each dose group and a blank control group. Spore suspensions from each dose group were evenly spread onto PDA agar plates and incubated in the dark at 23°C for 6 days. Colony counts were then recorded, survival rates were calculated, and dose-survival rate curves were plotted. The concentration of 1.5 × 10⁻⁶ cells corresponding to a 25% survival rate was determined. 14 ions / cm 2The midpoint of the sublethal dose window was used to induce three rounds of cumulative mutagenesis on the starting strain. After each round of mutagenesis, surviving strains were collected to prepare spore suspensions again, ultimately obtaining a strain library containing 256 effective mutants.

[0047] Next, machine learning-assisted quantitative surrogate phenotypic screening was conducted. With "high yield and tolerance to low temperature of 15℃" as the breeding objective, the culture temperature was controlled at 15℃ (the conventional culture temperature is 25℃) as the selection pressure. The strains from the mutant strain library were inoculated into conventional PDA medium and cultured in the dark at 15℃. Colony images were collected daily from the 4th day after inoculation using a 10-megapixel imaging device. Two quantitative surrogate phenotypic traits, colony edge regularity and hyphal density, were extracted using image analysis software. The target trait label was "hyphae growth rate ≥0.25cm / d at 15℃". The extracted phenotypic data were divided into training and test sets in a 7:3 ratio. A support vector machine classification model was constructed. After the model was trained, the accuracy of the test set was 80%. The surrogate phenotypic data of all mutant strains were input into the model, and 68 primary candidate strains were selected.

[0048] Subsequently, single-spore isolation, karyotype identification, and genetic stability verification were performed. Due to the low spore yield of *Flammulina velutipes* F47, single spores were directly picked and cultured using a micromanipulation method to obtain single-spore strains. Pure homokaryotic strains were screened through double karyotype identification, and the absence of clamping junctions was observed under an optical microscope. Specific primers were designed for PCR amplification targeting the mating type site of *Flammulina velutipes*. Only one band appeared in the amplification product, and 29 pure homokaryotic strains were finally obtained. Different mating type homokaryotic strains with low-temperature tolerance potential were backcrossed with the starting strain. After mycelial fusion, three rounds of subculture were performed (7 days per round). Mycelial growth rate and low-temperature survival rate were measured at 15℃ in each round. After three rounds of subculture, a total of 22 strains with a coefficient of variation of target traits <5% were identified as advanced candidate strains.

[0049] Finally, a multi-generation cultivation screening was conducted in a simulated industrial environment. Using a conventional industrial cultivation substrate (cottonseed hulls: wheat bran: corn flour = 75:20:5, moisture content 63%), 22 advanced candidate strains were inoculated into standard cultivation bags to build a simulated production environment. The humidity was controlled at 90%, the light cycle was 12h light / 12h dark, and the diurnal temperature fluctuation was set at 3℃ (e.g., 18℃ during the day and 15℃ at night). The advanced candidate strains were continuously cultivated for 3 generations. The yield of fresh mushrooms per bag at 15℃ and the survival rate in the low-temperature environment were measured for each generation. Finally, one low-temperature tolerant and high-yielding strain, F47-1, was selected, which met the requirements. Its average yield per bag over 3 generations was 0.92kg, and the survival rate in the low-temperature environment was 85%.

[0050] The existing control group uses a traditional chemical mutagenesis breeding method, specifically as follows: A spore suspension of *Flammulina velutipes* strain F47 was prepared and treated with 0.5 mg / mL nitrosoguanidine solution for 30 min (a commonly used high-intensity chemical mutagenesis parameter), resulting in 36 low-temperature tolerant mutants. These mutants were inoculated into a 15°C culture environment, and colony growth was observed manually. Based on experience, 18 candidate strains were selected and directly hybridized. After hybridization, only one round of subculture was performed before inoculation into cultivation bags. One generation of cultivation was carried out under undisturbed conditions of constant 20°C, 90% humidity, and a 12h light / 12h dark light cycle. The F47-control strain was selected, with an average yield of 0.82 kg per bag at 15°C and a low-temperature survival rate of 72%.

[0051] The experimental data of this embodiment group and the prior art control group were summarized, and the results are shown in Table 2.

[0052] Table 2 Comparison of Experimental Data Results

[0053] It should be noted that the industry standard data is based on the following: (1) The average yield per bag at 15℃ is 0.70kg. The yield of the 15℃ low temperature culture group (465.69g / 1200mL cultivation bottle) in "The Influence of Temperature and Time on the Yield of Enoki Mushrooms" is calculated by referring to the standard cultivation bag with a filling amount of 1.2kg and converting it into the industry standard yield; (2) The survival rate of 65% in the low temperature environment is derived from the optimal growth temperature (8-12℃) of Enoki mushroom fruiting bodies in "Biological Characteristics of Enoki Mushrooms". 15℃ is a low temperature stress environment. The survival rate of conventional unbred strains at this temperature is generally in the range of 60%-70%; (3) The relevant data also conform to the production performance description of conventional strains in the low temperature environment in "Temperature Segmentation Control Technology in Factory Production of Enoki Mushrooms".

[0054] According to the results in Table 2, the number of low-temperature tolerant mutants in this embodiment group was 2.1 times that of the control group through multiple rounds of cumulative mutagenesis induced by low-energy ion beams, effectively improving genetic diversity. The accuracy of screening using a support vector machine model combined with quantitative surrogate phenotypic screening was 22 percentage points higher than that of the control group, and the screening efficiency and accuracy were significantly better than manual screening. After single-spore isolation, karyotype identification and subgeneration verification after backcrossing, the trait stability rate was 40 percentage points higher than that of the control group, effectively reducing the risk of degradation of the low-temperature tolerance trait through generations. Multigenerational cultivation simulating industrial-scale low-temperature fluctuation environment resulted in a 31.4% increase in yield of the selected strain at 15℃ compared to the starting strain, which was higher than the 17.1% increase of the control group. The low-temperature survival rate reached 85%, which was better than the control group and the industry standard level.

[0055] Therefore, by adopting the above-mentioned breeding method for edible fungi, the genetic diversity of mutants can be greatly broadened, the accuracy and efficiency of target trait screening can be improved, the genetic stability of target trait can be ensured, and the adaptability of strains to industrialization environment can be enhanced. This can significantly improve the conversion efficiency of breeding results to large-scale production, and ultimately cultivate excellent edible fungi strains with excellent comprehensive agronomic traits that meet the requirements of industrial mass production.

[0056] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0057] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for breeding an edible mushroom, characterized by, The method comprises the following steps: S1, performing physical mutagenesis treatment on the starting strain of edible fungi, and based on the dose-survival rate relationship determined in the pre-experiment, performing multiple rounds of low-intensity cumulative mutagenesis within the sublethal dose window to obtain a mutant strain library with rich genetic background; S2, applying selection pressure associated with target traits to the obtained mutant strain library, extracting quantitative features as proxy phenotypes based on colony images collected during cultivation, and using a machine learning model to analyze the proxy phenotypes and screen out primary candidate strains; S3, performing single spore isolation and karyotype identification on the screened primary candidate strains, obtaining homokaryons, and performing controlled pairing or backcrossing on homokaryons with target traits to obtain advanced candidate strains with stable genetic background; S4, based on the obtained advanced candidate strains, performing multi-generation cultivation under conditions simulating actual production and containing environmental disturbances, and screening final strains meeting industrialization requirements based on yield and trait stability as indicators.

2. The method of claim 1, wherein the method is a method for breeding an edible mushroom, characterized by In S1, the physical mutagenesis treatment is selected from one or more of normal temperature and pressure plasma, low-energy ion beam, or closed ultraviolet mutagenesis.

3. The method of claim 1, wherein the method is a method for breeding an edible mushroom, characterized by In S1, the dose-survival rate relationship determined by pre-experiment is realized by the following method: Set up gradient mutagenesis dose to treat the starting strain, draw the dose-survival rate curve, and take the dose interval corresponding to the survival rate of 20%-40% as the sublethal dose window.

4. The method of claim 1, wherein the method is a method for breeding an edible mushroom, characterized by In S2, the proxy phenotype includes one or more of colony diameter daily growth, colony edge regularity, and mycelial density.

5. The method of claim 1, wherein the method is a method of breeding an edible mushroom, characterized by In S2, the machine learning model is a regression or classification model constructed using support vector machine or random forest algorithm.

6. The method of claim 1, wherein the method is a method of breeding an edible mushroom, characterized by, In S3, the single spore isolation is realized by plate dilution method or micromanipulation method, and the karyotype identification is realized by microscope observation of lock-like joint or PCR amplification detection using molecular markers for mating type sites.

7. The method of claim 1, wherein the method is a method of breeding an edible mushroom, characterized by In S3, after the controlled pairing or backcrossing, the obtained strains are further subjected to not less than three rounds of subculture for verifying the transmission stability of target traits.

8. The method of claim 1, wherein the method is a method of breeding an edible mushroom, characterized by In S4, the environmental disturbance includes diurnal temperature fluctuation of 1-3℃.

Images