About influenza A (H1N1) in terms of programming

Scientists have already completely disassembled the H1N1 and brought it to the NCBI Influenza Virus Resource virus database . There everything is documented in detail. For example, sample A / Italy / 49/2009 (H1N1) was found in the nose of a 26-year-old woman who returned from Italy to the United States. Here are the first 120 bits of its genetic code .

atgaaggcaa tactagtagt tctgctatat acatttgcaa ccgcaaatgc agacacatta

How many bits will kill a person?
According to rough estimates, the total size of the H1N1 source codes is 26,022 bits, and if you exclude service brake lights (indicate the end of each protein sequence), then the executable code consists of approximately 25,054 bits. This number is approximate also because the virus has a mechanism for generating excess debris for masking against antiviruses.
')
So, it turns out about 25 kilobits or 3.2 kilobytes. This is the amount of code for a program that has non-zero chances to kill a person. H1N1 is much more efficiently written than the MyDoom computer virus, about 22KB in size.

It is very humiliating that only 3.2 KB of genetic data can kill me. However, in 850 MB of the human genome, there must be holes for a couple of exploits.

For those not familiar with molecular biology, a short educational program. The symbols from the above code are called nucleotides. They are of four types (A [T / U] GC), that is, each character contains two bits.

The source code of a living organism (DNA nucleotide sequence) is loaded into RAM (one-to-one RNA nucleotide sequence is transmitted, except T is replaced with U), compiled as coding trinucleotides (codons) and launched for execution (codons synthesize amino acids of a specific organism ).

When the RNA program is launched for execution, the synthesis of amino acids begins. They are created by one amino acid for every three nucleotides. The above genetic code can be represented as such a sequence of basic amino acids.

MKAILVVLLYTFATANADTL

Each character in this entry (codon) is equivalent to six bits of information (three two-bit nucleotides). M is methionine, K is lysine, A is alanine, etc. (see amino acid table ).

Interestingly, the genetic code of any living organism begins with the start codon M. In nature, this is a mandatory marker of a valid code.

Since DNA and RNA are practically the same thing, in biology it is common practice to write any genetic code as a DNA nucleotide sequence, even if in nature this code is distributed directly ready for launch, that is, as RNA, like an influenza virus. This is a very important detail, which we will discuss below.

However, back to the main topic topic. The code given is the beginning of the HA gene, which programs the production of the hemagglutinin protein. Hemagglutinin of the H1 subtype is present in our sample, which is reflected in the name of the H1N1 virus.



Hemagglutinin is an interface for attaching to a host cell. If we imagine living creatures in the form of computers, then each functional group of cells in the body makes contact with the outer worlds through a certain port. For example, port 21 is specifically for FTP. In the influenza virus, port H1 indicates a specific area of ​​the human body (throat) to which you can join. Curiously, the same H1 is suitable for attaching to the intestinal tract of birds. Therefore, the same H1N1 virus is capable of attacking the human respiratory system and the birds' digestive system.

For comparison, another H5 port (part of the H5N1 virus, known as bird flu) is aimed at attacking lung tissue, which leads to pneumonia, so H5N1 is much more dangerous and causes much more deaths. Defeat through the port of H1 is not so scary: a person just starts blowing her nose a lot, but her lungs do not suffer.

Scientists are still studying the H5 port. It is already known that in some mutant people the lungs are not able to connect with the virus on this port. Who has such a mutation in the genes, has a great chance to survive the H5N1 epidemic without serious consequences if it suddenly begins.

Breaking swine flu
In the journal Nature published a fantastic article , which contains all the information about the current structure of the H1N1 virus, its comparison with other strains of influenza. The author describes in detail how the pathogenic component differs in them, that is, the part that kills a person.

With this information, I now know exactly how to modify the H1N1 source to make it more deadly.

The article says that the gene PB2 with glutamic acid in the 627th position has a weak pathogenicity (not very lethal). However, the lysine variant in the same position is more lethal. Well, we find this place in the H1N1 source. Information from the database:

601 QQMRDVLGTFDTVQIIKLLP
621 FAAAPPEQSRMQFSSLTVNV
641 RGSGLRILVRGNSPVFNYNK

As you can see, at the 627th position of the virus code is the symbol “E”, this is the codon for glutamic acid. Thank God that it is here that is why not so many people die from H1N1 as they could.

If we decode the codons into the DNA source code, we get:

621 FAAAPPEQSR
1861 tttgctgctg ctccaccaga acagagtagg

As can be seen, glutamic acid corresponds to the nucleotide sequence “GAA”. To modify the genome and make the virus more deadly, we only need to replace “GAA” with one of the two lysine variants (“K”), that is, “AAA” or “AAG”. Thus, a more lethal version of H1N1 will look like this:

621 FAAAPPKQSR
1861 tttgctgctg ctccaccaaa acagagtagg

Like this. Swap only two bits - and the almost harmless H1N1 turns into a more deadly version.

Theoretically, I can use quite affordable technologies to synthesize a new virus. For a start, I can contact one of the commercial companies that are engaged in DNA synthesis (for example, Mr. Gene ) and order a specific DNA sequence for less than $ 1,000. True, you need to consider that the company Mr. Gene applies a screening procedure to identify a potentially dangerous biological code. It is difficult to say whether their filters will react to this modified version of the gene. If so, then you can do the production of a new virus and without their help. To do this, we need to get samples of the usual H1N1 and again apply the well-known and standard techniques for controlled mutation of the genome to replace the desired nucleotide.

By the way, Nature even has a link to a scientific article , which details how to make a virus of influenza A on your own. Proc. Natl Acad. Sci. USA 96, 9345-9350 (1999)]. Also an interesting read.

Resourceful flu
Before we modify the H1N1 virus, let's think about it. Mother Nature did a brilliant job by packing the death code into just 3.2 KB, so much so that we can't handle it. Apparently, she is not so stupid. Our little hack won't be a revelation to her. Maybe the flu can change a couple of bits in its own code?

The short answer is yes.

The influenza virus is indeed capable of such mutations. The fact is that after copying the DNA molecule, the protein is launched to check the "checksum". He verifies that the copy is identical to the original. But the problem is that the influenza virus is based on RNA with a proprietary copying mechanism. It does not run a checksum check at all. As a result, the level of copying errors is extremely high: about one per 10,000 base pairs. And this is despite the fact that the entire flu genome consists of 13,000 base pairs. That is, roughly rounding off, there is one random mutation in each new copy of the flu.

Some of these mutations are irrelevant, others make the virus harmless, but some very rare mutations can make the virus more dangerous. Since viruses multiply and spread at an astronomical rate, the chances that our little hack happens naturally are very high.

This is one of the reasons, I think, why medical organizations are so concerned about the spread of H1N1: we have no protection against it, and although it is not very dangerous so far, after a couple of mutations, really serious problems can begin.

Currently, the pathogenicity of H1N1 is about the same as that of normal flu. In June, there were only 87 deaths in 21,449 confirmed cases, that is, the mortality rate is 0.4%. For comparison, the “normal” flu has a mortality rate of <0.1%, in the Spanish variety in 1918 the mortality rate was 2.5%, and in the H5N1 (bird flu) variety, the mortality rate exceeds 50% (!), But Fortunately, almost not transmitted from person to person.

Having infected H1N1 today, you can get a valuable bonus of practically safe immunity, so that when the virus mutates and comes back again, you will be under protection.

Another problem with the influenza virus copying system is that it stores its genetic code in eight sections of RNA, and not in one indivisible chain. Because of this, the replication mechanism becomes even more unpredictable. If you are “lucky” to get infected with two different flu viruses, then a fundamentally new strain of flu may appear in your body, because flu viruses in your infected cells exchange code in a chaotic manner.

This is why the additional danger of H1N1 lies in its unique property, which is called “triple reassortment”. It contains fragments of human, and swine, and bird flu, and can freely exchange the code with its "relatives" of any kind of animals.