The technique of using drugs to target latent virus reservoirs has been introduced to reawaken dormant viruses so that the immune system can attack them. However, further tests have shown this method to fail in laboratory tests. In this work, the author tries to mathematically analyze whether drugs can be used to reawaken dormant virus reservoirs and proposed the use of viral proteins to activate the sleeping virus. The results show that the amino acid sequences ARG of gag proteins of HTLV1, HTLV2, STLV1 and STLV2 match with their primer binding site GGGGGCTCG in the 3'-to-5' direction, and the amino acid sequences SPR of gag proteins of HIV1, HIV2, SIV and FIV match with their primer binding site GGCGCCCGA in the 3'-to-5' direction. The gag proteins are promising for reawakening dormant retrovirus infection. The author hence believes that the latency-reversing drugs were involved in the process of transcription of cancer genes, and the virus genome they reawaken were just happened to contain the same NF-κB binding sites, so the drugs were indirectly reawakened dormant retrovirus infection. On the other hand, it is more reliable to use viral proteins to directly reawaken retrovirus, just as androgen receptor activates the IGF1R gene.