machine learning – mycolab.ai

Decoding of novel umami-enhancing peptides from Hericium Erinaceus and its mechanisms by virtual screening, multisensory techniques, and molecular simulation approaches

mycolabadmin

Researchers discovered four special proteins (peptides) from lion’s mane mushrooms that can enhance the savory umami taste of foods while potentially allowing for less salt in products. These peptides work by helping salt compounds stick better to taste receptors in your mouth. This discovery could help food companies create healthier products with better flavor but lower sodium content, reducing the health risks associated with excessive salt consumption.

XenoBug: machine learning-based tool to predict pollutant-degrading enzymes from environmental metagenomes

mycolabadmin

XenoBug is a new artificial intelligence tool that helps scientists find bacteria and their enzymes that can break down harmful pollutants like pesticides, plastics, and petroleum products. The tool analyzes genetic information from environmental samples to predict which enzymes can degrade specific toxic chemicals. This discovery approach could make environmental cleanup faster and cheaper by identifying the right microbes for the job. Researchers can use XenoBug to get starting points for developing new biological cleanup solutions.

A novel dataset of annotated oyster mushroom images with environmental context for machine learning applications

mycolabadmin

Researchers have created a large collection of carefully labeled photographs of oyster mushrooms along with environmental data from the farm where they were grown. The dataset includes about 16,000 images showing mushrooms at different stages of growth, captured both day and night, along with measurements of temperature, humidity, and air quality. This resource is designed to help scientists and farmers develop computer programs that can automatically identify mushrooms, determine if they’re ready to harvest, and predict growth patterns.

Multilayer perceptron-genetic algorithm as a promising tool for modeling cultivation substrate of Auricularia cornea Native to Iran

mycolabadmin

Black ear mushrooms (Auricularia cornea) are nutritious and medicinal fungi that can be grown on waste materials from wood industries. Researchers tested different combinations of sawdust and bran to find the best growing mixture. They used artificial intelligence to predict which combinations would give the best yields, finding that a mix of 70% beech sawdust with 30% wheat bran worked best and could be produced efficiently.

Harnessing carbon potential of lignocellulosic biomass: advances in pretreatments, applications, and the transformative role of machine learning in biorefineries

mycolabadmin

This comprehensive review examines how agricultural and forestry waste containing lignocellulose can be transformed into valuable products like biofuels, packaging materials, and medical supplies. The paper covers various treatment methods to break down the tough plant material structure and highlights how artificial intelligence can improve these processes. By utilizing this abundant waste resource efficiently, we can reduce environmental pollution, generate renewable energy, and create useful products while supporting a circular economy approach.

Natural-selected plastics biodegradation species and enzymes in landfills

mycolabadmin

Landfills contain billions of tons of plastic waste that can take centuries to decompose naturally. This research discovered that landfill microorganisms have evolved to break down plastics through natural selection. Using advanced computer analysis of microbial DNA, scientists identified thousands of potential plastic-degrading enzymes that could be engineered for industrial applications to help clean up plastic pollution.

Exploring neural markers of dereification in meditation based on EEG and personalized models of electrophysiological brain states

mycolabadmin

Researchers developed a new brain-monitoring technique called the Inner Dereification Index that can detect when someone is meditating versus mind-wandering using only a brief EEG recording. By analyzing electrical activity in specific brain regions involved in self-awareness and personal thoughts, the method can accurately track meditation progress in real-time with 99.6% accuracy. The technique works with minimal training data and shows that certain meditation practices—particularly Tibetan Buddhist techniques aimed at experiencing the emptiness of self—create distinctive brain patterns. This breakthrough could enable real-time meditation feedback devices and personalized meditation guidance.

Biomass carbon mining to develop nature-inspired materials for a circular economy

mycolabadmin

This paper explains how we can turn waste biomass from agriculture and industry into valuable materials to replace petroleum-based products. By using computational methods and artificial intelligence, researchers can design more efficient processes to convert plant and animal waste into bioplastics, chemicals, and building materials. Over 100 companies are already successfully doing this, creating products from waste coffee grounds, seaweed, agricultural residue, and other biomass sources.

Mushroom data creation, curation, and simulation to support classification tasks

mycolabadmin

This study creates a new dataset of over 61,000 mushroom records from 173 species to help computers learn to identify whether mushrooms are safe to eat or poisonous. The researchers extracted mushroom information from an identification textbook and used computer programs to generate realistic hypothetical mushroom entries. They tested different AI methods and found that random forests (a type of machine learning algorithm) worked best, achieving perfect accuracy in identifying poisonous versus edible mushrooms.

Advancing automated identification of airborne fungal spores: guidelines for cultivation and reference dataset creation

mycolabadmin

Researchers developed standardized procedures to grow fungal spores in laboratories and prepare them for testing with automated detection devices. They tested 17 different fungal species commonly found in the air and created reference datasets to train computer algorithms to identify these spores. Two different detection technologies were evaluated, showing promising accuracies (55-95%) for identifying various fungal spores. This work provides a blueprint for other scientists to create reliable training data for automated air quality monitoring systems that track allergens and disease-causing fungi.

Research Topic: machine learning