RT-PCR Testing for Software Engineers

What is RT-PCR?

RT-PCR stands for Reverse Transcription Polymerase Chain Reaction. Some literatures declares RT as Real-Time. These two are different but produces the same output.

Polymerase Chain Reaction, from what I learned (thank you CHEM32 Chemistry of Biomolecules), is a technique used to amplify copies of a portion of the polynucleotide sequence in a sample of DNA. Polynucleotide is a sequence of nucleotide base pairs. You’ll remember polynucleotides from your high school biology class: the A, G, C, T. To amplify means to make a copy of this DNA into hundreds to thousands of copies.

Then what is RT? The central dogma, to put it simply, states that DNA makes itself (through replication), DNA makes RNA (through transcription) and RNA makes proteins (through translation). The reverse does not, or rarely, happen normally. Since RT means Reverse Transcription, it means that RNA is used to make DNA.

Philippine Genome Center recently made a webinar for everyone’s perusal.

How is RT-PCR done?

The video above perfectly describes the process of RT-PCR in a molecular level. However, since we’re not really a graduate of Biology, I’ll simplify this a bit.

Steps:

1. If a specimen has no DNA, extract RNA.
2. If it has DNA, using restriction enzymes, cleave DNA into smaller fragments. For RNA, combine the extract with appropriate mix of enzymes and primers.
3. For RNA, start initial amplification.
4. Start cycle:

• Denaturation — heat specimen at around 95°C to split DNA strands into one strand
• Annealing/Priming — mixture is cooled down at around 60°C to allow primer to bind to the strand
• Synthesis/Extension — polymerase enzyme creates the complement of the strand, starting adjacent to the primer

Step 4 can be done multiple times, depending on the needs. Repetition increases the yield amount of target sequence and homogeneity.

Are we the lab technicians? Why discuss it?

No, as software engineers, we’re not the ones doing the laboratory analysis but that does not mean we can’t help. What if we design an information system based on the use cases?

1. The user can add, edit, and deactivate users.
2. The user can add, edit, and delete roles. The user can also assign roles to users.
3. The user can add, edit, and retrieve patient information.
4. The user can add laboratory input data.
5. The user can add laboratory analysis output data.
6. The system can compute, interpret, and generate a result.
7. The user can verify and approve/disapprove a generated result.
8. The system can notify the patient of their results.
9. The patient can access their record securely.
10. The system can generate a print out of the result.
11. The system can share and contribute data to the open biomedical databases. (optional)

What are the important details that an information system should take?

Let’s break down the data from the laboratory.

• Patient Data — who owns the specimen; full name, sex, birthday, classification
• Requestor Information [optional] — depending on your current set-up, if you are a laboratory who takes requests from other institutions, this might help in the identification; name, address, laboratory identifier
• Specimen Details — what was given to the laboratory personnel; type, accession number or identifier, date and time collected, date and time received
• Laboratory Result Details — unless the manufacturers give lab machine APIs for the developers, the only way to record the lab results are through human interventions;

For the laboratory result details fields, we’ll base this with the information system for MERS-CoV used by a laboratory here in the Philippines.

1. Cycle Threshold (CT) values: Sample upE, Sample N2, Positive Control upE, Positive Control N2
2. Positive Control Batch Number
3. NA Extraction Kit Used
4. PCR Machine Used

The fields above may change depending on the user’s needs. The genes to test for varies.

Patient information is trimmed down to only the important matters. In reality, the Department of Health requires a four-page form for the data reporting of laboratories. Depending on your country, your information system must adhere to any laws that tackles data privacy.

Interoperability is still not in the top priority. However, in designing, keep in mind that as enterprise architecture grows, the systems needs to keep up with communication.