In the ongoing severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic hamster species have been particularly valuable to study host-pathogen interactions as they are permissive to develop a moderate (Mesocricetus auratus) or severe (Phodopus roborovskii) disease course following infection. Here we use single-cell ribonucleic acid sequencing of fresh whole blood, lysed of erythrocytes, to dissect cell-specific changes to the transcriptome in the early stages of moderate and severe corona virus disease 2019 (COVID-19) in SARS-CoV-2-infected hamsters. To determine species-specific transcriptional responses, our generated datasets were integrated with publicly available transcriptomic COVID-19 patient data featuring moderate and severe disease courses. Analyses were performed using R package Harmony as well as Python packages Scanpy and scGen, which enable disease state predictions using an autoencoder neural network architecture. For this, a low dimensional latent space embedding was used to capture the most relevant transcriptome data and identify shift vectors from healthy to diseased cells. Preliminary results show that the interspecies integration of hamster and human data is achievable and all major cell types can be characterized throughout the datasets. Training of a variational autoencoder enables to map hamster COVID-19 temporal disease states to human COVID-19 severity ranks for individual cell types. Additionally, the established workflows could subsequently be used to study the pathology of extensive lung diseases, shedding light on cellular mechanisms in the transition from healthy to diseased cellular states.