This paper describes results of comparison of genetic sequences of influenza viruses obtained from clinical samples (nasopharyngeal swabs) and viruses isolated on MDCK cells and in developing chick embryos. The obtained data shows that 1-2x fold passaging of influenza viruses in MDCK cells does not lead to additional mutations in the genome of the influenza virus, and the genetic material of this virus is identical to the genetic material from the clinical sample. When passaging viruses in chick embryos within 1-2 passages, the genetic material of the virus differs from the genetic material from the clinical sample, because of mutations in the receptor region of hemagglutinin, leading to aminoacid substitution Q223R. The mutation Q223R changes the receptor specificity of hemagglutinin from human-type receptors (α-2.6-sialic acids) to avian receptors (α-2.3-sialic acids). The data obtained coincide with the data of Japanese researchers who showed that a similar substitution in the hemagglutinin gene was detected in Japanese strains of influenza virus subtype A (H1N1) pdm09, passaged in chick embryos for 1-2 passages. With their further passaging in chick embryos, the number of acquired mutations in the surface proteins of these viruses increases. Thus, we have shown that for the genetic analysis of epidemic-relevant strains of influenza viruses for routine epidemiological surveillance tasks, the use of direct sequencing of clinical samples containing the genetic material of viruses makes it possible to obtain the most reliable data, to increase their efficiency, since it saves time normally spent on virus isolation procedure. If preliminary isolation of viruses is necessary, for example, if it is not possible to carry out full genome amplification from a clinical sample (low content of the material, presence of polymerase chain reaction inhibitors), virus isolation on MDCK cells with a minimal amount of passage is optimal, which is used in practice of epidemiological surveillance.