Results
To train HEATMAP-AI for enzyme temperature prediction, we collected and organized data from public databases and scientific literature, which will be a valuable resource for future researchers.
To download the data, please visit Website.
Additionally, to make our dataset more accessible, we have created a platform where users can easily search for and download enzyme data.
To access the database, please visit our Website.
To evaluate the performance of our model, we utilized the cross-validation method. This robust statistical technique assesses how well the results of a model generalize to an independent dataset. Below are the results obtained from our cross-validation process.
The Pearson correlation coefficient of both are up to 0.89, which proves that our model not only has the advantage of high accuracy, but also has excellent robustness, which indicates that our model has very strong universality.
On top of this, HEATMAP-AI is used to build etcGEM model. The successful construction of etcGEM also means the model is excellent.
In the wet-lab section, we validated the accuracy of the HEATMAP model's predicted optimal temperature for enzymes by investigating the catalytic efficiency of trypsin in hydrolyzing casein at different temperatures. Based on the amino acid sequence of trypsin, the HEATMAP model predicted an optimal temperature of 43.71°C. Consequently, we designed three reaction systems at 33.5°C, 43.5°C, and 53.5°C, respectively. The catalytic efficiency under these temperature conditions was characterized by measuring the absorbance of the reaction product, L-tyrosine, at its characteristic absorption peak of 275 nm, using a Thermo Scientific NanoDrop spectrophotometer to monitor the trypsin cleavage of α-casein.
The absorbance values of the 43.5°C reaction group at wavelengths of 270 nm, 275 nm, and 280 nm were 0.243, 0.285, and 0.283, respectively, which were higher than those of the 33.5°C reaction group (0.221, 0.265, 0.270) and the 53.5°C reaction group (0.236, 0.277, 0.280). This likely indicates that more L-tyrosine was produced at 43.5°C, suggesting that the catalytic efficiency of trypsin was highest at this temperature, potentially corresponding to its optimal temperature. This observation aligns with the HEATMAP model's predicted optimal temperature of 43.71°C, thereby validating the HEATMAP model's accuracy through wet-lab experiments.
To better explore the key enzymatic systems involved in the production of Spinosad by \( Saccharopolyspora\:spinosa \), we constructed an enzyme-constrained genome-scale metabolic model (ecGEM) using mathematical modeling techniques, integrating \( k_{cat} \) deep learning predictions and protein pool flux constraints to characterize its overall metabolic network.
Figure A shows the simulation of exchange fluxes at various growth rates using the ecGEM model.
Figure B illustrates the proportional usage of protein pools in the ecModel as the growth rate changes for the final constructed ecGEM model.
Figure C presents the proportional usage of protein pools in the ecModel as the growth rate changes for the ecGEM model without adjusted protein pool constraints.
Figure D depicts the proportional usage of protein pools in the ecModel as the growth rate changes for the ecGEM model without adjustments to kcat values.
The results demonstrate that the final ecGEM model adequately simulates the normal physiological state of \( Saccharopolyspora\:spinosa \) and progressively achieves a matched proportion of protein usage corresponding to the growth rates.
Here, you can click the corresponding link to download our constructed ecGEM model.
This model can be optimized further or applied to investigate the specific functions of the \( Saccharopolyspora\:spinosa \) metabolic network.
To access modeling data results, including DLKcat predicted kcat values and ecGEM simulations of the growth state of \( Saccharopolyspora\:spinosa \), visit the Model section.
In order to explore the effects of temperature on biological growth. Based on the ecgem model in 2022 SJTU-software, we constructed etcGEM model with HEATMAP-AI and DARWINS in 2023 SJTU-software.
The simulation shows that the growth situation differs in different temperature. What's more, It can be observed that the most suitable growth temperature for the bacterial strain is around 56 degrees Celsius. When the temperature rises or falls, the activity of its internal proteins decreases, thereby affecting the growth and development of the entire organism.
Here, you can click the corresponding link to download our constructed etcGEM model.
Figure 1: HEATMAP Prediction of the Optimal Working Temperature for Proteinase K. Absorbance Results of Enzyme Reaction Systems at Different Temperature Gradients: We conducted full-wavelength absorbance scans on the supernatants from the reaction systems at 33.5°C, 43.5°C, and 53.5°C. Notably, all three groups exhibited a significant absorbance peak between 250 nm and 290 nm, with a peak position around 275 nm. This corresponds to the characteristic absorption peak of L-tyrosine, indicating that Proteinase K successfully cleaved α-casein in all three experimental groups.
Figure 2: Full Wavelength Absorbance Scan of the Supernatant from the Enzyme Reaction System at 33.5°C, covering the range from 190.0 nm to 850.0 nm.
Figure 3: Full Wavelength Absorbance Scan of the Supernatant from the Enzyme Reaction System at 43.5°C, covering the range from 190.0 nm to 850.0 nm.
Figure 4: Full Wavelength Absorbance Scan of the Supernatant from the Enzyme Reaction System at 53.5°C, covering the range from 190.0 nm to 850.0 nm.
Absorbance Values at 270 nm, 275 nm, and 280 nm under Different Temperature Gradients: The enzyme reaction at 43.5°C showed absorbance values of 0.243, 0.285, and 0.283 at 270 nm, 275 nm, and 280 nm, respectively. These values were higher than those observed at 33.5°C (0.221, 0.265, 0.270) and 53.5°C (0.236, 0.277, 0.280). This strongly suggests that the 43.5°C reaction group produced more L-tyrosine, indicating that Proteinase K likely exhibits its highest catalytic efficiency at this temperature, which may represent its optimal temperature. This finding aligns with the HEATMAP model's predicted optimal temperature, thereby validating the model's accuracy based on the wet experimental data.temperature and confirming the model's accuracy through wet experiments.
Figure 5: Comparison of Absorbance at 270 nm, 275 nm, and 280 nm in the Supernatants of Enzyme Reaction Systems at 33.5°C, 43.5°C, and 53.5°C.