INTRODUCTION

The principles of engineering can be applied to solve synthetic biology problems. Once we identify the problem, our engineering cycle begins with designing a solution that seems feasible. In the build stage, we extend the idea and check for the feasibility of the solution. Next, we test if the solution we came up with will work. In this manner, we learn the improvements to be made, and the cycle continues.

Figure 1. The Design-Build-Test-Learn (DBTL) cycle

Our project ‘AptaSteles’ stands as testimony to the cycle. At every phase, the project underwent multiple iterations of the Design-Build-Test-Learn (DBTL) cycle. Integrating expert opinions, we reformed our system to achieve a better result.

In this section, the following are highlighted.

Biomarker identification and analysis: After considering multiple factors like the source, ease of extraction, and reliability, it was decided that the biomarkers from the blood, such as microRNAs (miRNAs), proteins, and hormones, would be the best options for detection.
miRNA detection: Because the miRNA levels are very low, we included a procedure for amplification. After an extensive literature review, we settled on the isothermal miRNA-Recombinase Polymerase Amplification (miRPA) technique to amplify miRNAs. Subsequently, the Fluorescent Aptamer Sensor for Tracking microRNAs (FASTmiR) module was used for detection.
Protein and Hormone detection: The concentrations of biomarker proteins and hormones are also very low. Therefore, the signal given by the aptamers should be amplified, and a dual aptamer approach was adopted for this purpose.
Kit Design: As the detection system changed, so did the kit design, to accommodate them. Finally, a 3D prototype of a microfluidic chip was printed and tested. The optical components were then accordingly optimised.

Modules

Module 1: Biomarker Analysis

Initially, we identified that the gene DENND1A is overexpressed in PCOS cases. The variant V2 mRNA of the same is specific to PCOS and can be detected in urinary exosomes [1]. Considering this to be our biomarker, we progressed to a stage where we tried to build upon its detection mechanism. However, on interacting with the iGEM 2021 team IISER-Tirupati_India, we realised that PCOS, being a syndrome, cannot be concluded based on one biomarker. So we started our search for several biomarkers from different sample sources.

Aim: Finding biomarkers from a given sample source.

First cycle:

Design: Keeping DENND1A V2 as our basis, other biomarkers in urine were searched for. Literature review revealed the existence of testosterone glucuronide and 11-α hydroxyprogesterone in urine [2], which were known to indicate PCOS. According to Dr Shunmathe K R, the estrogen forms’ E1 and E2 ratio: E2/E1 is also a biomarker used by medical professionals. These were hence chosen as the biomarkers and further worked upon.

Build: As details of the biomarkers were looked upon, it was found that DENND1A V2 was excreted out of the body, packed in exosomes (vesicles from cell membrane that carries cargo).

Test: Various mechanisms for the uptake and internalisation of the exosomes for the mRNAs to be detected were assessed. Cell-penetrating peptides and cell wall remodelling were a few of the options considered for the uptake of exosomes. Internalisation mechanisms involving the expression of CD81 receptors were found suitable [3-5]. Since eukaryotic proteins were to be expressed and functionalised, the host organism chosen was Pichia Pastoris, a model yeast strain.

Learn: We learned that specific information about the exosomes and the markers expressed on them were required to internalise the same. However, there was not enough details about them. Also, the biomarkers chosen were not inclusive of all the symptoms. We selected the biomarkers that indicated the changes happening in the ovaries; however, they did not give much idea about the other metabolic disorders in the body. The chosen biomarkers could vary due to other factors like the consumption of birth control pills and the usage of steroids, to name a few. Thus they cannot determine specifically PCOS. Considering the complications of internalisation and the compromise on accuracy of the chosen biomarkers, we left the idea of detecting the given ones from urine, Thus, we moved on to the next cycle.

Second cycle:

Design: The search for biomarkers continued. We got to know that many biomarkers in the blood are indicative of different symptoms associated with PCOS. A meeting with Dr Surbhi Singh gave us an idea about the various symptoms considered clinically for diagnosing a person with PCOS. So the biomarkers were selected in accordance with that, along with the other potential biomarkers in consideration, to ensure the kit's specificity towards testing PCOS.

Build: 10 potential biomarkers were identified from the blood. The biomarkers class ranges from microRNAs to proteins and hormones. (Click Blueprint to check for the individual biomarkers). The biomarkers were selected so that many comorbidities associated with PCOS could be detected.

Test: The biomarkers were checked for their specificity and sensitivity toward PCOS from existing literature. We found the specificity and sensitivity values for the chosen biomarkers. It is observed that the biomarkers had relatively high sensitivity towards PCOS (Click Blueprint to check for the individual values). So with this data, we successfully identified the biomarkers. We quickly found that detecting these biomarkers could cover a range of associated comorbidities, thus serving our purpose.

Learn: We learned that the combinations of a few of our biomarkers yielded better results. So we envisaged a model to find the best combination of biomarkers. But Dr Shibdas Banerjee pointed out that we need our experimental results for the same. Due to time constraints and ethical issues, we could not proceed with it.

Figure 2. The chosen biomarkers

Figure 3. Summary of module 1

Module 2: miRNA detection

Among the biomarkers to be detected, four of them were miRNAs. Since miRNAs are only 22-25 bases long, they could be unequivocally complementary to any sequence. So the choice of a proper detection mechanism was tough. After consideration of the pros and cons of the different mechanisms for detecting miRNAs, it was found that the toehold system is the best suited for the project.

Aim: To find the technique for the detection of miRNAs

First cycle:

Design: The usual detection technique used for mRNAs is riboswitches. We modified it to use microRNAs.

Build: A riboswitch circuit is shown in the diagram below. We mined the database for the miRNA sequences to build the genetic circuit. In this riboswitch, the trigger sequence is the miRNA, and the toehold is the antisense of the miRNA. The gene of interest is GFP or any other fluorescent protein.

Figure 4. Representation of Riboswitch

Test: We tried validating the circuit by finding different factors like the length of the stem loop and the length of toehold that could affect the protein production. So we checked how these could be done in our specific case with the miRNAs of interest.

Learn: As we were developing a point-of-care device, we learnt that the employment of protein-based detection could make our system less reliable due to the labile nature of proteins. The protein fluoresces only on maturation (post-translational modifications). This leads to a time lag for the fluorescence to appear. In a protein based detection system, degradation of the protein is inevitable, which hinders the accuracy of the output it can give. Since we aim to quantify the biomarkers, we learnt that the protein-based detection method is not ideal. So we tried to avoid adding translational machinery to our system.

Second cycle:

Design: Moving away from protein based systems, the next option we discovered is aptamers (Click Background to know more about aptamers). Aptamers offered many advantages over protein systems in terms of cost, adaptability, stability, and many others. So with aptamers as the basis, we looked for methods of detecting miRNAs. We came across Fluorescent Aptamer Sensor for Tracking microRNAs (FASTmiR) used in detecting miRNA 122 [12]. The FASTmiR is an RNA sequence that is an extension of aptamers. It utilises light-up aptamers and a complementary sequence of the miRNA of interest (Click Blueprint to know more about FASTmiR).

Build: We found the FASTmiR approach to be suitable for our system and extended it to our biomarkers. Before testing, it was necessary to know quantitatively the amount of miRNA found in the blood.

Test: There was no data about our specific biomarkers. However, it is known that the total concentration of miRNA in cells is about 10^-12M [13]. Based on basic estimates done using Ct values of the miRNAs, the concentration of miRNA was assumed to be in the range of 10^-14 M.

We tried to model this system using rate law kinetics. A graph of the concentration of FASTmiR-dye complex formed as a function of the concentration of miRNA (that we calculated to be present in the blood) was plotted.

Figure 5. Graph from FASTmiR model

Learn: The graph shows that the time taken to reach an equilibrium concentration of the aptamer-dye complex is in the range of 10¹⁸ s. We could immediately infer that the reaction was happening very slowly. This inference matched the data we got from the calculation, proving that the miRNA level is too low to be detected by FASTmiR.

Third cycle:

Design: From the previous cycle's results, we realised the need for an amplification technique to increase the concentration of the miRNA biomarkers. We explored isothermal amplification techniques as they should be incorporated into the kit.

Build: After an extensive literature survey and vigorous brainstorming, we settled on miRNA Recombinase Polymerase Amplification(miRPA), inspired by the iGEM 2018 team City of London, UK (Click Blueprint to know more about miRPA) [14-15].

Test: We developed a mathematical model to understand the system and find the concentration after amplification.

Figure 6. Graph from miRPA model without looping

The graph clearly showed that there was a very minimal amplification happening. However, the fold change was maintained nearly as the initial fold change. Then, as Dr Jatish Kumar pointed out about chain kinetics, we accounted for the recycling of miRNA and duplexed miDNA and looping back into the system.

Figure 7. Graph from miRPA model with looping

We succeeded in achieving a significant quantity of biomarkers in 30 minutes.

Learn:

As seen in the graph, the fold change is still maintained. Theoretically, it was hypothesised that the same would be reflected in the concentration of the FASTmiR-dye complex.

Fourth cycle:

Design:

We wanted to test our hypothesis. So the DNA sequences were designed in the following manner. The light-up aptamer used was modified Spinach, an RNA light up aptamer. The sequence on transcription has a binding pocket that particularly binds to DFHBI dye (a fluorophore) and gives fluorescent output (Click Model to know more about the designs).

Figure 8. Design of FASTmiR with terminator

Build:

Figure 9. Graph from FASTmiR model with amplified product

A mathematical model was built to check the concentration of FASTmiR- dye complex formed with the input being the amplified miDNA (output from miRPA) (Click Model to know more about the model). The fold change was reflected in the model.

Test:

While designing our sequences, we gained input from various people. We then tested those sequences and checked their functionality. We did PCR amplification for the sequences to see that we were getting multiple bands. It was inferred that due to the length increment with the T7 terminator, the primers may have had more chances of non-specific binding, due to which we got multiple bands.

Figure 10. Test of FASTmiR-146a all designs. One can see multiple bands of varying lengths.

Learn:

We learnt that there should not be a terminator at the end. The presence of the terminator poses the risk of non-specific primer binding and non-specific/incorrect IVT product.

Fifth cycle:

Design: Meanwhile, we ordered our sequence-specific reverse primers that would ideally remove the T7 terminator from transcription and increases the specific binding of primers, lowering the chances of getting multiple bands.

Figure 11. Design of FASTmiR without terminator

Build: We used our sequence-specific primers for PCR amplification and got the following result:

Figure 12. Gel showing fewer faint multiple bands with the new reverse primers

Figure 13. Successful IVT with desired band length showing the removal of T7 terminator.

Figure 14. Fluorescence readout for miR-146a-D4 after 30 mins incubation.

Figure 15. Fluorescence readout for miR-146a-D4 after 1 hour incubation.

Figure 16. Fluorescence readout for miR-222-D2 after 30 mins incubation.

Figure 17. Fluorescence readout for miR-222-D2 after 1 hour incubation.

The validated IVT product shows fluorescence that matches the expected modelled fold change.

Figure 18. Summary of Module 2

Module 3: Protein and Hormone Detection

Aptamers are used widely for different types of molecules ranging from small molecules to proteins. So we decided to use aptamers to detect proteins and hormones because of their versatility.

Aim: To find an appropriate detection mechanism using aptamers for proteins and hormones.

First cycle:

Design: We reviewed the different ways aptamers are raised for a given ligand. There are methods like Systematic Evolution of Ligand by EXponential enrichment (SELEX), De Novo Rapid In Vitro Evolution of RNA biosensors (DRIVER) and in silico methods using softwares like Making Aptamers Without SELEX (MAWS) developed by the iGEM 2015 Team Heidelberg.

Build: We searched for characterised DNA/RNA aptamers specific to the protein and hormone biomarkers. DNA aptamers for Luteinizing Hormone, Testosterone, CRP were identified. However, two of the protein biomarkers, Prolidase and Irisin, do not have any well-characterised aptamers for them. So, the detection plan bifurcated into two:

For the ones where there are no aptamers, we would raise the aptamers through SELEX.
The ones we have aptamers for would be used for biomarker detection.

Test: We purified proteins Prolidase and Irisin for SELEX. We also consulted experts to find ways to optimise the SELEX protocol.

Figure 19. Gel showing the elutes of Prolidase

Learn: SELEX protocol could be optimised to reduce the number of cycles and the time between each cycle.

Second cycle:

Design: We designed our SELEX experiments based on the lesson learnt from the previous cycle and expert advice.

Build: We made a standard protocol for DNA selex and developed an outline of experimental flow that needs to be followed.

Test:

Figure 20. Gel image showing incorrect library

We tested to verify whether our received orders i.e, our library (79 nucleotides and the primers) are purified and of the desired size.

Learn: The DNA library we received was wrong.

Third cycle:

Design:

Since the library we got was wrong and there was a time constraint to order it again. So we looked for ways by which we could come over it. Then we realised the existence of an in-house DNA/RNA synthesiser.

Build:

We synthesised our own library of sequences using the protocol we standardised for ourselves.

Test :

We tested if the obtained library is the one required by running it on 10% native gel.

Figure 21.Gel showing the correct synthesised library

From the results, it was verified that the library is right. Thus we successfully synthesised our library. However, due to time constraints, we could not move forward with SELEX.

Fourth cycle:

Design:

For the other plan of detection, we came across the usage of light-up aptamers for visualising non-coding RNA in live cells (Click Background to know more about light-up aptamers) [10]. The concept of light-up aptamers gave us the start to build the system.

Build:

This system was built in parallel to SELEX using the already characterised aptamer.

The aptamers are fused with a light-up aptamer. When the target ligand of the aptamer binds to it, change in the conformation of the light up aptamer is induced. This leads to the binding of the dye to our aptamers, thus leading to fluorescence.

Figure 22.Fusion of light up aptamer and biomarker specific aptamer

Test:

The factors that affect the binding of the ligand to the aptamer were checked. It was found that truncating the extra regions of the aptamer, changing the distance of three-way regions, and many others could enhance the binding. We tried drawing comparisons to existing designs and testing the factors in our system.

Learn:

We realised that the biomarker concentration was too less to give a detectable fluorescent output. Quantifying the biomarkers based on such a low fluorescence output would be very difficult, and neither would it give a precise result.

Fifth cycle:

Design: Since the fluorescence output was very low, there was some kind of amplification required. Unlike miRNAs which could be amplified before detection, for proteins and hormones, alternatively, the signal had to be amplified. Taking inspiration from [17], we devised the dual aptamer approach.

Build: The biomarker-specific aptamer has a trigger sequence bound to it. On binding of the biomarker, a conformational change is observed, which releases the trigger. The trigger sequence binds to the target sequence, which linearises and exposes the gene coding for light-up aptamers. This leads to the transcription of light-up aptamer which binds to the dye giving rise to a fluorescence complex (Click Blueprint for more information about the approach).

Test: We designed the aptamer-trigger complex for 2 of the selected biomarkers (Testosterone and C-reactive proteins and 2 other biomolecules (Progesterone and Quinine) that could help us achieve our proof-of-concept. We built a mathematical model to understand the formation of the light-up aptamer complex. (Click Model for more information about the designs and the model)

Figure 23. Graph from dual aptamer model for hs-CRP

Figure 24. Graph from dual aptamer model for testesterone

Learn: We understood the working of the system and found that the light up aptamer reached an equilibrium after 3.5 hours. We went about testing the sequences experimentally.

Sixth cycle:

Design: After the conceptual designing of the system, we designed two dual aptamer approaches to detect proteins and hormones.

Approach 1: With phi29 DNA polymerase
Approach 2: Without phi29 DNA polymerase

Each dual aptamer system consists of biomarker-specific aptamer, trigger and target. The system is based on the release of triggers from the aptamer upon biomarker binding to the aptamer. This trigger release ultimately leads to the amplification of the signal by transcription of the light-up aptamer, Broccoli. The amount of biomarker is correlated with the fluorescence intensity of Broccoli at a given point in time.

Build: Using in-silico tools like NUPACK, Unafold, DuplexFold and analysing the thermodynamical parameters of the aptamers, we made the pipeline that followed to design of these triggers. (CLick Model to know more). We designed multiple triggers for a given aptamer. For each biomarker, we made 3-7 trigger designs of varied lengths.

Test: Upon experimentally testing various designs for each approach, we realised that using approach 2 (without phi29 DNA polymerase), the system failed to detect any hormonal concentration, while approach 1 (with phi29 DNA polymerase) was successful in detecting the hormone at a given concentration.

Figure 25. Workflow to test our designs for approach 1

Our experiments showed promising results for the testosterone dual aptamer system

Incubation of testosterone with T6-D3

Figure 26. Fluorescence readout for T6-D3 at various concentrations of Testesterone

The above figure indicates that the concentration of testosterone has no effect on the fluorescence readout.
Incubation of testosterone with T6-D1 and D2

Figure 27. Fluorescence readout for T6-D1 and D2

The above figure clearly shows that for D2 (Design 2) for the testosterone aptamer, the system was able to sense 1mM of testosterone and give a significant fluorescence readout which is 4 folds higher than the control (no hormone).

Part number

Part name

BBa_K4438600

Aptamer_T6

BBa_K4438603

T6_trigger_2_phi29

BBa_K4438604

T6_Target_2

We also tested all designs for progesterone and quinine aptamers, but we could not identify a working design. However, the design for T6-D2 would be a successful template for optimisations of designs in the future.

Learn:

From testing of all the plethora of designs, we realised the following learnings:

Complementarity between trigger and aptamer plays a very crucial role in both annealings of trigger-aptamer and release of trigger upon biomarker binding to aptamer.

Optimisation of aptamer-trigger concentrations could help increase the sensitivity of the system.

A one-pot reaction for Approach 2 (Without phi29) is not achievable due to the higher affinity of the trigger to the target. Hence, Approach 2 could be more prone to failure.

This led to optimisations in our kit design.

Figure 28. Summary of Module 3

(Comment after the competition) NOTE:

After troubleshooting our results, we realised that “Approach 1” wouldn’t be feasible due to the incorrect polarity of the sequences. In theory, this approach is not possible.

We realised that “Approach 2” is the only feasible solution and this approach could be modified to detect proteins and hormones involving phi 29 DNA polymerase.

The suggested modifications in “Approach 2” are:

Design the trigger sequence (removing sense T7 promoter sequence at 3’ end) complementary to an arbitrary sequence of the biomarker-specific aptamer.

The trigger sequence can have a few bases complementary to the antisense T7 promoter sequence at the 3’ end. Sequentially added phi 29 polymerases (after strand displacement) will form a double-stranded DNA promoter to initiate the transcription by T7 RNA polymerase.

Parameters like free energy of binding and secondary structures of all possible trigger-target sequences and trigger-aptamers sequences should be verified. This is necessary in order to maintain the specificity of binding in the presence and absence of the protein/hormone/ligand.

Module 4A: Kit design

Since our project was about making a diagnostic kit, we felt it was incomplete without the development of hardware. We immediately realised the potential of microfluidics. Thus we went about building a device based on the principles of microfluidics. We changed our designs to minimise the wastage of sample and to incorporate sequential addition of the reagents. To get a better picture of the iterations our kit design underwent, please check Hardware

Aim: To optimise the microfluidic chip design and the components.

First cycle:

Design:

Our microfluidic chip design underwent multiple iterations. With the help of Dr Dileep Mampallil, we came up with 7 designs (Click Hardware to know more about the designs). Each design was an improved version of the previous one.

Build:

The designs were scrutinised based on the feasibility, ease of replication, and complexity. We then fixed two among them, one of them is the implemented kit and one to show a preliminary proof of concept.

Figure 29. Printed microfluidic Chip

Figure 30.a Before Illumination

Figure 30.b After Illumination

Test:

We printed a mould of the design using a 3D printer and fabricated a Polydimethylsiloxane (PDMS) chip using it. The chip was made hydrophobic using teflon. It was tested with water.

The water from the storage chamber moves to the reaction chamber on gently pressing the storage chamber. Next we coated a new chip of the same design with silane (more hydrophobic). We tested the system with dye and broccoli light-up aptamer by adding them to the respective chambers. We successfully observed fluorescence on pressing the chamber containing the dye.

Module 4B: Signal Processing

Since the output from our system is fluorescence, it was a necessity to have a proper sensing mechanism in place that could detect the fluorescence in the most accurate way.

First Cycle:

Design:

Based on the requirements, the different components of the circuit were decided upon. One among them was the selection of the right mechanism for processing the signal. There were different options for the same.

Process the signal to give the electrical output to an Arduino board coded to give the digital output.

Convert the current output to voltage with a trans-impedance amplifier. The voltage output is then converted to digital output by an analog-to-digital converter MCP3302/04 MCP3426/7/8 .

The current output is converted to digital output with a DDC264/DDC316 Current-Input Analog-to-Digital Converter ,which combines the current-to-voltage and A/D conversion.

In a discussion with Dr Kanagasekaran and Mr Sivaraju, we came to the conclusion that the first option might be the best way out.

Build:

We built the circuit in the following way to test the photodiodes and the Arduino's functionality.

Figure 31. Circuit Setup and Validation

Test:

We developed the code for the Arduino.

Initially, we got either a “High” or “Low” display, which was dependent on the threshold. But it didn't change as we increased or decreased the light intensity value as expected.

Initial code:

https://static.igem.wiki/teams/4438/wiki/engineering-success/initial-code-for-arduino-2.pdf

Thus we changed the commands multiple times, and were successful in the end.

Final code which gave the result

https://static.igem.wiki/teams/4438/wiki/engineering-success/final-code-for-arduino.pdf

Figure 32. Summary of Module 4

REFERENCES

McAllister, J. M., Modi, B., Miller, B. A., Biegler, J., Bruggeman, R., & Legro, R. S. (2014). From the Cover: PNAS Plus: Overexpression of a DENND1A isoform produces a polycystic ovary syndrome theca phenotype. Proceedings of the National Academy of Sciences of the United States of America, 111(15), E1519. https://doi.org/10.1073/pnas.1400574111

Wang, W., Wang, S., Tan, S., Wen, M., Qian, Y., Zeng, X., ... & Yu, C. (2015). Detection of urine metabolites in polycystic ovary syndrome by UPLC triple-TOF-MS. Clinica Chimica Acta, 448, 39-47.

,Ruseska, I., & Zimmer, A. (2020). Internalization mechanisms of cell-penetrating peptides. Beilstein journal of nanotechnology, 11, 101–123. https://doi.org/10.3762/bjnano.11.10

Rizzo, J., Rodrigues, M. L., & Janbon, G. (2020). Extracellular Vesicles in Fungi: Past, Present, and Future Perspectives. Frontiers in Cellular and Infection Microbiology. https://doi.org/10.3389/fcimb.2020.00346

Gurung, S., Perocheau, D., Touramanidou, L., & Baruteau, J. (2021). The exosome journey: From biogenesis to uptake and intracellular signalling. Cell Communication and Signaling, 19(1), 1-19.

Bhatnager, R., Nanda, S., & Dang, A. S. (2018). Plasma prolidase levels as a biomarker for polycystic ovary syndrome. Biomarkers in Medicine, 12(6), 597-606.

Zhang, L., Fang, X., Li, L., Liu, R., Zhang, C., Liu, H., Tan, M., & Yang, G. (2018). The association between circulating irisin levels and different phenotypes of polycystic ovary syndrome. Journal of endocrinological investigation, 41(12), 1401–1407.

Sumithra, N. U., Lakshmi, R. L., Leela Menon, N., Subhakumari, K. N., & Sheejamol, V. S. (2015). Evaluation of Oxidative Stress and hsCRP in Polycystic Ovarian Syndrome in a Tertiary Care Hospital. Indian journal of clinical biochemistry : IJCB, 30(2), 161–166.

Abdelazim, I. A., Alanwar, A., AbuFaza, M., Amer, O. O., Bekmukhambetov, Y., Zhurabekova, G., Shikanova, S., & Karimova, B. (2020). Elevated and diagnostic androgens of polycystic ovary syndrome. Przeglad menopauzalny = Menopause review, 19(1), 1–5.

Sørensen, A. E., Wissing, M. L., Salö, S., Englund, A. L. M., & Dalgaard, L. T. (2014). MicroRNAs related to polycystic ovary syndrome (PCOS). Genes, 5(3), 684-708.

Long, W., Zhao, C., Ji, C., Ding, H., Cui, Y., Guo, X., ... & Liu, J. (2014). Characterization of serum microRNAs profile of PCOS and identification of novel non-invasive biomarkers. Cellular Physiology and Biochemistry, 33(5), 1304-1315.

Huang, K., Doyle, F., Wurz, Z. E., Tenenbaum, S. A., Hammond, R. K., Caplan, J. L., & Meyers, B. C. (2017). FASTmiR: an RNA-based sensor for in vitro quantification and live-cell localization of small RNAs. Nucleic acids research, 45(14), e130-e130.

Max, E. A., Bertram, K., Akat, K. M., Bogardus, K. A., Li, J., Morozov, P., Ben-Dov, I. Z., Li, X., Weiss, Z. R., Azizian, A., Sopeyin, A., Diacovo, T. G., Adamidi, C., Williams, Z., & Tuschl, T. (2018). Human plasma and serum extracellular small RNA reference profiles and their clinical utility. Proceedings of the National Academy of Sciences of the United States of America, 115(23), E5334. https://doi.org/10.1073/pnas.1714397115

https://2021.igem.org/Team:City_of_London_UK

Wee, E. J., & Trau, M. (2016). Simple isothermal strategy for multiplexed, rapid, sensitive, and accurate miRNA detection. Acs Sensors, 1(6), 670-675.

Paige, J. S., Nguyen-Duc, T., Song, W., & Jaffrey, S. R. (2012). Fluorescence imaging of cellular metabolites with RNA. Science, 335(6073), 1194-1194.

Sheng, L., Lu, Y., Deng, S., Liao, X., Zhang, K., Ding, T., ... & Li, J. (2019). A transcription aptasensor: amplified, label-free and culture-independent detection of foodborne pathogens via light-up RNA aptamers. Chemical Communications, 55(68), 10096-10099.

Namasivayam, V., Lin, R., Johnson, B., Brahmasandra, S., Razzacki, Z., Burke, D. T., & Burns, M. A. (2003). Advances in on-chip photodetection for applications in miniaturized genetic analysis systems. Journal of Micromechanics and Microengineering, 14(1), 81.

Part number	Part name
BBa_K4438600	Aptamer_T6
BBa_K4438603	T6_trigger_2_phi29
BBa_K4438604	T6_Target_2