Team:NCTU Formosa/HP/Gold Integrated
Industrial Circles
FWUSOW INDUSTRY CO., LTD.
FWUSOW INDUSTRY CO., LTD. Is one of the well-known pesticide manufacturers. In recent years, FWUSOW INDUSTRY got the ball rolling of the bio-pesticide production and had a place on the research of biopesticide application.
Leon Liang - Deputy Manager
Suggestion:
“ Your platform of peptide prediction is pretty prospective and innovative and can even overthrow the current ways to find the new function of peptides. The platform could be better if you give users more details about the environmental conditions which the pathogens live. “
Figure 1: Photo with the managers of FWUSOW INDUSTRY. The manager Leon Liang is the first person on the left and the manager Kevin Chen is the first person on the right
Kevin Chen – Deputy Section Manager
Suggestion:
“ The peptide prediction system is needed to the pre-period of bio-pesticide researches and development. It could significantly save time and money of the processes. And the system can try adding more functions, such as connecting to the database of Central Weather Bureau. In that case, it is able to show the local environmental information of where the user is via GPS.”
Figure 2: Make a presentation to introduce our project
Government Agency
Taoyuan District Agricultural Research and Extension Station, COA
Taoyuan District Agricultural Research and Extension Station is a national government agency set up for Agricultural Science and Technology development and committed to improving traditional agriculture via high-tech engineering.
CHANG,WEI-PIN - Technical Specialist of Plant Epidemic Prevention Lab
Suggestion:
“Current preventions to pathogens are not effective and eradicative, because of a forecast of where the pathogens appear is rough without accuracy. Government agencies could only tell when did the disease prevail in the past by data with experiences and ask the farmers to be alert. That’s not a prediction! How come a prediction ranges of weeks or months? So, the IoT system with the pathogen erupts forecast is an urgent need to a perfect prevention, if you can also combine the IoT forecast prediction system and your Parabase, the whole system could be complete, useful, and competitive!”
Figure 3: Visiting expert. Photo with the professor CHANG,WEI-PIN in Taoyuan District Agricultural Research and Extension Station
WU,YOU-HENG – Researcher of Biomechatronics Lab
Suggestion:
“The concept of using AI to do the prediction to the function of peptides and the occurrence of diseases is really an amazing and outstanding idea. You build the bridge between peptide applications to agriculture and the disease forecast, which could be pretty useful for recent agriculture. I think you could visualize the peptide prediction score, which might be easier for people to understand, that will make your system more user-friendly.”
Figure 4: Visiting expert. Photo with the professor WU,YOU-HENG in Taoyuan District Agricultural Research and Extension Station
Taiwan Agricultural Chemicals and Toxic Substances Research Institute, COA
TACTRI is a national institute responsible for pesticide research and development, the monitor of pesticide residues and toxic substances in agricultural products, plant protection technology development, pesticide technical services, and the establishment of analytical methods and evaluation standards to ensure pesticide rational usage and the safety of agricultural products.
Feng-Chia Hsieh- Associate Researcher and Head of Biopesticides Division
Suggestion:
"It is outstanding and creative to combine artificial intelligence and drug selection, that costs down much time on searching antifungal peptides. Apart from fungi, diseases caused by viruses and bacterias are also delicate and serious in agriculture. In your description, you told me that SCM could also be applied to train more types of data, so I’ll be looking forward to seeing a powerful and complete system that can identify antiviral and antibacterial peptides. The disease occurrence model with IoT truly brought the intelligence to the applications of agricultural hardware, it will be a big contribution after farmers introduce it in a real farm. For the experiment part, if there’s more time for the project, you can do experiments with more types of pathogens.”
Figure 5: Visiting expert. Photo with the professor Feng-Chia Hsieh in Taiwan Agricultural Chemicals and Toxic Substances Research Institute. Dr. Hsieh is the person on the left.
Academic Circles
Department of Plant Pathology and Microbiology, National Taiwan University
The department committed to learning the interaction between plant pathology and microbiology, which is one of the most powerful fungal research institute in Taiwan.
Ying-Lien Chen - Professor
Research domain:
- Antifungal drug development
- Integrated management of potato common scab
- Genetics and pathogenesis of Candida and Fusarium
Suggestions:
Now the government promotes the idea of Agriculture 4.0, which aims to combine the power of artificial intelligence and big data analysis of Internet of Things. Sad to say, not many people can do this, but your team, gives a push to the concept.
Your team combines IoT and use it to predict the probability of disease occurrence, it’s an outstanding work, and by the way, it would be more convenient to use biopesticides. A few suggestions for you, apart from the environmental condition you have already concerned, you can add more factors while training your data, that can fit the condition better. For instance, the microclimate and biological specificity of ecology.
Also, to prevent antifungal diseases, you can try a way to help plants increase their resistance. Find some protein that can activate the immune system of plants, it also needs analysis of AI, and with these things optimized, that’ll be impressive.
Figure 6: Visiting expert. Photo with the professor Feng-Chia Hsieh in Taiwan Agricultural Chemicals and Toxic Substances Research Institute. Dr. Hsieh is the person on the left
Department visit and Academic exchange Dr Frank Yates - CellTechs Laboratory, Sup’Biotech
Researches in Modeling neurodegenerative diseases with induced pluripotent stem cells (ips) and neuroectodermal organoids (mini-brains) Research Axes:
Automation of biochemical techniques and cell culture.
Development of a system of eucaryotic and procaryotic expression for the production of recombinant antibodies (ScFv).
Suggestions:
“It is creative to predict the specific function of a protein by using the algorithm analyzing the primary structure. And the Scoring Card Method is truly an interpretable way for users to analyze the features of amino acid sequences with the idea of dipeptides. The project would have a huge contribution to the drug selection for the agriculture and even the pharmaceutical industry. For further work in experiments, you can do a deeper analysis on the protein degradation, because the degradation is an important issue which decides the impact of peptides.”
Figure 7: Group photo with the Dr Frank Yates, who is the second person on the right
Conclude
After getting lots of advice from various fields, we optimized and adjusted the Parabase website to be more user-friendly.
1. Easier to browse
To catch user’s eyes and to allow users to be familiar with this tool immediately, we beautified the website surface, simplified the searching steps and added instructions.
2. Easier to understand and interpret
People might not know the meaning of the predicted scores, so we turned the number into color codes and presented it on the corresponding sequences. Also, we explained how did the score and the threshold be determined and the meanings of them.
3. Easier to search
We added the function of searching common names in host searching apart from searching binomial names. Most users like students or farmers will need one more step to transfer the common names into binomial names because they are not familiar with binomial names of plants, so we are now solving this inconvenience by keeping adding common names in Parabase.
Parabase Website
(It is the Home page of our Parabase Website)
