Enzymes play important roles in the biological production of synthetic biology. By introducing various enzymes into the reaction systems and biological devices, we are able to produce many valuable products in mild external conditions or within the living organisms. However, a complete biosynthetic or degradation pathway contains multiple ‘nodes’ and contains multi-step reactions. In the process of research, we sometimes find that the enzyme which catalyzes a certain step of the pathway has a low efficiency, or even there is no readily suitable enzyme in the database to catalyze the target reaction, which brings some difficulty to the study of metabolic pathway. Fortunately, enzyme confounder has been shown to be universal, that is to say, many enzymes have the ability and potential to catalyze other reactions in addition to specificity, which comes from the naturally existing metabolic regulatory mechanism in organisms. Although this ability to catalyze reactions different from original ones can be difficult to retrieve in commonly used databases, we can still search by reaction similarity comparison.

In the preliminary research stage of determining our topic of project, we had an in-depth exchange with several professors and students engaged in synthetic biology research in the college. During the conversation, one professor kept complaining about the time-consuming and repetitive experiments when screening for an enzyme in the research. We talked a lot with him about the problem. Naturally, we had such thoughts below and gradually came up with our project idea.

iGEMers and synthetic biologists frequently complain it repeated and time-consuming to screen manually for an enzyme when there is no available suitable ones in the database to catalyze the target reaction, as we investigated. They regularly do such work simply based on experience or maybe just by random attempt. Based on this problem, we conducted further research and communication. We found that several existing software aims to solve these problems. For example , XTMS applies inverse synthesis to search for the metabolic pathways that produce the target compound; and PathPred, proved useful in multistep response prediction. Yet, they do have several defects. For example, XTMS, based on a limited E. coli database, could not identify the precursor material; and using PathPred, it is still inevitable that intermediate reactions will be disconnected.

MEI, modified enzyme interface, is a user-friendly platform that was designed specifically for synthetic biologists to solve such issues above. MEI, literally the pinyin for enzyme in Chinese, it can predict the enzymes required for a single step reaction, score the results based on a series of criteria and output in a user-friendly way. Therefore, it can shorten the experimental time and find the desired enzyme more quickly.

By using our software, in a very short time, users can obtain a series of candidates for potential enzyme that is most likely to catalyze the target reaction, with scores to describe the likelihood of the relevant catalysis. Despite this,. And our software can output the enzyme activity data, with auxiliary reactions taken into account. Therefore, our enzyme prediction platform can greatly reduce the experimenters' needless searching time.

Generally, the workflow can be described as follows. First and foremost, by using our user-friendly input interface, users can easily draw or upload the structural formula of the substrate and product and mark the group that changed during the reaction. Then, users can select the cofactors and the species of enzymes he wants, as well as the target reaction type. After that, users can get the enzyme prediction results with their scores. In addition, users can obtain the recommended auxiliary reaction and its feed ratio.

Of course, users don't need to recite so many command codes. All you need may be a single mouse. Moreover, our software performs a better user experience through our simple and clear input interface.

Through these things, MEI can literally predict, I mean, predict the potential enzyme candidates to catalyze the target reaction. Surely, it can help synthetic biologists shorten the experimental time and find the desired enzyme more quickly.

MEI has a great visual input interface. You can use it easily, and you don't need to study different forms of chemical labeling.

MEI prioritizes the type of reaction required. You will not get irrelevant enzyme results.

MEI fully considers cofactors, enzyme species and enzyme activity. You can apply it directly to the real world.

1. Hult K, Berglund P. Enzyme promiscuity: mechanism and applications. Trends Biotechnol. 2007 May;25(5):231-8. doi: 10.1016/j.tibtech.2007.03.002. Epub 2007 Mar 26. PMID: 17379338.

2. Carbonell P, Parutto P, Herisson J, Pandit SB, Faulon JL. XTMS: pathway design in an eXTended metabolic space. Nucleic Acids Res. 2014 Jul;42(Web Server issue):W389-94. doi: 10.1093/nar/gku362. Epub 2014 May 3. PMID: 24792156; PMCID: PMC4086079.

3. Moriya Y, Shigemizu D, Hattori M, Tokimatsu T, Kotera M, Goto S, Kanehisa M. PathPred: an enzyme-catalyzed metabolic pathway prediction server. Nucleic Acids Res. 2010 Jul;38(Web Server issue):W138-43. doi: 10.1093/nar/gkq318. Epub 2010 Apr 30. PMID: 20435670; PMCID: PMC2896155.