@pemile
comparé 2 atomes a la grosseur d’un virus qui en contient plusieurs milliers
The smallest known animal-infecting virus is the porcine circovirus 1, with a single-stranded DNA genome of 1759 bases, as mentioned by Alec Cawley. If you believe that the genetic material wholly constitutes the virus, then, with an average of 35 atoms per nucleotide (base plus sugar plus phosphate), you get 60,000 atoms.
However, the physical form of the virus is a 17-nm diameter icosahedron — protein walls surrounding a water and DNA-filled core. The virus’s buoyant density has been measured to be 1.35 g/ml, which computes to a viral mass of 3.4 x 10^-18 g. The mass of the DNA is 1.0 x 10^-18 g of this. The mass of the water is likely very small, based on my intuition as a sometime biophysicist. So let’s assume that 2.4 x 10^-18 g mass is protein, and assume that the protein’s elemental composition is the same as the average for bacteria. (If I had infinite time, I would look up the capsid protein sequence and calculate this, but this is good enough.) The mean atomic weight for atoms in a protein is about 12.85. Putting all these numbers together, we get 120,000 atoms in the viral proteins and 60,000 in the genome, yielding 180,000 atoms of virus.
The term « virus » is not all that well-defined. There are certainly smaller self-replicating genetic elements — satellite viruses, ribozymes, transposons, etc. These all have a somewhat smaller degree of autonomy compared to the virus discussed here, but are certainly worth reading more about !