Why SARS-CoV-2 was ‘explained’ by a computer

~ 13 minutes reading time article | blog.tulpas.dev//2020/SARS-CoV-2-Analysis

A paper first published July 7th (as of 2020) reads “A mechanistic model and therapeutic interventions for COVID-19 involving a RAS-mediated bradykinin storm” with the only relevant content published alongside being a very short abstract.

Later on, most of their relevant findings were published on the paper’s official portal editing the brief abstract and replacing it with research findings able to explain key characteristics from SARS-CoV-2, keys features which remained attributed to randomness by most scientists due to how they linked by mechanisms that could not be traced back to a single factor.

Using pharmacology and genetics to explain how SARS-CoV-2 works is nothing new, and many respectable papers have tried diverse approaches to this theme, the most notables being:

Albeit these studies are relevant to the matter and discussing pharmacology and SARS-CoV-2 pharmacokinetics properties is one of the best approaches to unknown pathologies, they are biased towards explaining observable mechanisms and fail to identify underlying conditions that could be attributed to a single common underlying cause, seemingly only being able to explain what has been previously seen or studied, as a result their theories are limited on a theoretical framework.

Instead of focusing on their more orthodox approaches, I will respectfully diverge from their perspective as we’re dealing with a novelty pathogen. This article’s objective is to share a different perspective on an issue that seems to have a different face-value for most, as in, there is a a lot of subjectivity analysis being thrown in ways that undermine a less knowledgeble reader’s understanding. Said that, I’m not an expert on the medical field nor should any of what I will just say be cited unless you’re providing proper sources and context, and perhaps, valid criticism. You should not take medical advice or knowledge from strangers on the internet.

A brief historical academic review

Initially understood as a virus weaponized to affect our lungs, authorities quickly noticed something was amiss, shortly afterward a main binding mechanism was identified: angiotensin-2, or ACE-2 receptors, an unusual group with no known mechanism linked to cytokine storms nor respiratory diseases, coupled with these findings came the first wave of hospitalized patients and efforts in identifying the disease characteristics, where doctors noticed this virus had serious symptoms such as cytokines storms and deadly pneumonia with severe affected lungs as a result, where patients regularly needed artificial breathing (i.e. ventilators) in order to stay alive.

‘Cytokine storm’ is a loose term for classifying scenarios where your immunity system cells is unable to properly identify the foreign entity - such as a virus - and as a result it starts attacking its own tissues, a behavior that initially classified immune-depressed population and chronic respiratory diseases as higher “deadly risk” for currently infected SARS-CoV-2 patients. A conclusion that does not fit meta-analysis from historical data, as smokers were considered to be one of the most at-risk groups from initial analysis, instead, they turned out to have a lower hospitalization ratio than your average non-smoker, surprisingly (1, 2, 3, 4) a conclusion supported by data even when considering gender, age distribution, underlying conditions and race.

(Although I have to assume that subject is heavily controversial, as this study shows)

Instead, scientists started to notice most worrying stealthy characteristics of the virus were actually aimed at our cardiovascular system, as the Troponin I biomarker could be used to monitor how severe the infection was on hospitalized patients, a protein measured to infer how much damage is being generated to our heart tissues, that’s why it is commonly used as a signaler for medics to forecast and alert heart attacks. So anyone with chronic heart diseases was included in the risk group ahead of existing respiratory conditions, as cardiovascular comorbities started to surge as able to predict a patient outcome in terms of likelihood of survival. The ACE2 receptor and its interactions were and still are under heavy studies, since the mechanisms behind the virus interactions with the receptor, even when isolated, are not easy to determine the consequences in a complex environment such as a the human body. One key early concept worth noting is sourced from a study citing ACE2 enyzmes as ‘Trojan Horses’, explicitly focusing on the virus cardiovasuclar implications and understanding all pharmokinectics involved between our receptors and the virus.

A risk factor at first deemed irrelevant surged silently between medical community, as more data provided information about how damaging the virus was to our kidneys, one of the body’s organs most rich in ACE-2 receptors number counts, it seemed like an intuitive analysis: kidneys are main targets for the virus and a major risk factor. This is one of the key characteristics that really surprised me on how backward the initial analysis were if you would consider a ‘bradykinin storm’ scenario. As science progressed in understanding the virus, the consensus skewed towards a seemingly less understandable version of what the virus actually was, challenging even ‘common-sense’ facts as applying sciencetific methodologies and their inability to cope with the unknown spectrum of knowledge, forcing even the most knowledge person in the subject to high degrees of doubt (e.g. 1, 2).

A machine-made hypothesis

Aware of how an unknown mechanism was at play, Daniel Jacobson, a computational systems biologist at Oak Ridge National Laboratory in Tennesee, USA, reported something was strangely amiss back in April when looking at gene expresssion data from lung fluid of COVID-19 patients, he spotted that fluid balance overall, including blood pressure regulation was unusually askew, and the expression of key enzymes for the renin-angiotensin system, also known as RAS, was directly correlated with these observations, which led to further investigation into which mechanism could cause such a severe deregulation.

To understand how this fits into a larger scientific context, it is important to understand the protein known as Angiotensin-converting enyzme 2, or ‘ACE2 Receptor’ is the main target for SARS-CoV-2, which is a theory largely supported by the scientific community, and what is more striking, it is hypothesized in most papers is that virus acts like an inhibitor, meaning the virus act as a blocker for specific enzymes on the receptor it is attaching itself to, reducing expression of each receptor’s signaling, and as a result, provokes an inhibitory effect on our system’s targeted cells.

What has been discovered by Daniel Jacobson team through a super-computer called Summit, the world’s second largest and potent supercomputer, was a result from analyzing possible interconnected proteins related to COVID-19 infections, which results were astonishing from the scientific perspective, the computer’s findings implicated the mechanisms used by the virus were reducing ACE expression while increasing ACE2 expression, playing an unthinkable role, as there are scarce information about viruses acting as receptors agonists, mimicking the protein behavior in order to speed the cell’s production of enzymes.


The leaky blood vessels and lung fluid build-up in some COVID-19 patients might be explained by the virus’s corruption of an inflammation safeguard, namely, ACE2’s degradation of DABK. Read the full description of the pathway here.


From this short piece of information alone one is able to understand why such a simple view, that was largely not considered due to the lack of analogous examples in other known diseases and virology scientific knowledge overall had such a contrarian perspective. A hypothesis considering this pharmacological behavior is able to explain why dosing ACE2 inhibitors as antihypertensive agents did not pose a threat during the pandemic as expected by doctors early on, as antagonists will probably interfere in the virus behavior as they would be competing for receptors ligands and could pose no major difference to minor improvement in COVID-19 cases. Data seems to support that hypothesis, as the supposedly complications from combining taking this kind of medication with COVID-19 would see a drastic rise in hospitalizations, which meta-analysis studies didn’t see statistical significance on infected patients taking ACE2 medications (1, 2).

A series of alleged pharmacological implications can be derived from this mechanism’s behavior, e.g. if the virus acts as a ACE inhibitor and a with a behavior like a ACE2 agonist, you should see something similar to what dosing Angiontensin-II anti-hypertension inhibitors medication in overdose does to the human body with pharmacological knowledge indicating we should see problems similar problems to what have been observed on patients of this virus, such as: clotting problems, strange neurological complications, an increase of the production of hyaluronic acid - a material that fills patients lungs and can absorve a thousand times its weight in water - thyroid problems and related hormonal desregulation and cardiac and kidney serious issues. As to actual overdoses, rare observaations have been made by doctors pre-COVID era, but the ones reported strike with dreading similarities.

In blunt terms

The idea that a virus could be mimicking a mechanism of action that has never been previously observed opens the bioinforrmatics field for new possibilities, where exploratory analysis becomes mandatory in early scientific studies, and hopefully, improves therapeutics options for treating new and threatening diseases of all kinds.

Further reading: 1, 2, 3, 4.