Team:SJTU-Software/Proof Of Concept
{IGEM_TopBar}}
Theoretical support
The idea of our project is proved by the article Cancer diagnosis with DNA molecular computation[1], which have shown the feasibility of using miRNA as a target for early cancer screening through DNA computation.
To successfully aid the design of the probe, we have done a lot of research on DNA strand displacement reaction[2-6]. We conclude from these articles that the secondary structure of the probe and hybridization with other substance coexisting in the system are crucial factors in DNA computation process.
Dry lab validation
Data analysis
We have collected a great amount of miRNA expression data and integrated them into our project. Here, we use a dataset of lung cancer, which has the highest incidence among all cancers, to demonstrate the validity of our data analysis tools.
First, we perform a differential expression analysis on the data.
Next, we use the above miRNAs to construct and train the SVM classifier, from which we obtain two up-regulated miRNAs and two down-regulated miRNAs with the best classification effect, and obtain the best model.
It can be seen that we have successfully screened out miRNAs with statistically significant differential expression. And the classifier constructed from these miRNAs yields an accurate prediction of the health condition.
Secondary Structure Prediction
We use the test set to measure the predictive ability of our model：
Model F1 score and accuracy:
The boxplots show that our model is feasible and robust for secondary structure prediction.
Wet lab validation
The smooth progress of DNA computation process
Dr. Qian Ma has researched in the field of DNA computation. This is a picture of the fluorescence signal she obtained in the wet-lab experiment. In the process of realizing DNA computation, the design of the probe has utilized all the tools provided by our project. The smooth progress of the reaction proves the validity of the probe and further proves the reliability and effectiveness of our software.
Assist probe design in other wet-lab experiments
We have been in close contact and cooperation with the SJTU-BioX-Shanghai team. In part of their projects, they are aiming to add two probes to both ends of the aptazyme. Normally, the aptazyme has the tendency of self-cleaving. In their experiments, the probes mentioned are essential for following procedures, but they shouldn't affect the original structure of the aptazyme. So they need to make sure that the probes added will not form hydrogen bond with bases in the functional domain.
This is the predicted secondary structure of aptazyme with probes added on both end. (The circled parts are the probes. The predicted structure shows that they won't affect the original structure.)
This is the electrophoresis picture of the experiment, the upper band is the content of the uncleaved part and the lower band is the content of the cleaved part.
The contents of the four lanes are (From left to right):
- Lane 1: Stop oligo is added to the system which affect the aptazyme and can hinder its self-cleaving function
- Lane 2: A kind of small molecule that can hinder the function of aptazyme is added (without stop oligo) .
- Lane 3: Without substance that can hinder the function of aptazyme.
- Lane 4: Start probe is added to recover the function of aptazyme hindered by stop oligo.
We can see from Lane 2 and Lane 3 that the function of aptazyme has not been affected (can be cleaved normally and suppressed by small molecule). It proves that the original structure are not affected by probes on both ends, which shows the effectiveness of our secondary prediction tools.
The predicted secondary of start probe:
Lane 1 and Lane 4 provide information about the stop oligo and start probe. The start probe doesn't receive a good effect, which is probably because that it will form a secondary structure by itself. This phenomenon conforms with the conclusion given by our prediction tool.
Reference
[1] Zhang, C., Zhao, Y., Xu, X. et al. Cancer diagnosis with DNA molecular computation. Nat. Nanotechnol.15,709–715 (2020). https://doi.org/10.1038/s41565-020-0699-0
[2] Simmel FC, Yurke B, Singh HR. Principles and Applications of Nucleic Acid Strand Displacement Reactions. Chem Rev. 2019 May 22;119(10):6326-6369. doi: 10.1021/acs.chemrev.8b00580. Epub 2019 Feb 4. PMID: 30714375.
[3] Zhang D.Y. (2011) Towards Domain-Based Sequence Design for DNA Strand Displacement Reactions. In: Sakakibara Y., Mi Y. (eds) DNA Computing and Molecular Programming. DNA 2010. Lecture Notes in Computer Science, vol 6518. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-18305-8_15
[4] Zhang DY, Winfree E. Control of DNA strand displacement kinetics using toehold exchange. J Am Chem Soc. 2009 Dec 2;131(47):17303-14. doi: 10.1021/ja906987s. PMID: 19894722.
[5] Song T, Garg S, Mokhtar R, Bui H, Reif J. Analog Computation by DNA Strand Displacement Circuits. ACS Synth Biol. 2016 Aug 19;5(8):898-912. doi: 10.1021/acssynbio.6b00144. Epub 2016 Jul 22. PMID: 27363950.
[6] Simmel FC, Yurke B, Singh HR. Principles and Applications of Nucleic Acid Strand Displacement Reactions. Chem Rev. 2019 May 22;119(10):6326-6369. doi: 10.1021/acs.chemrev.8b00580. Epub 2019 Feb 4. PMID: 30714375.