Fig. 5
Patient-associated drivers of plasma virome structure. A Estimated fixed effects of linear mixed effect regression on plasma cluster 3 (LMM, p < 0.05). Patient ID was used as explanatory variable, cluster abundance as response, and time after transplantation entered as random factor. The fixed effect estimates were colored according to their sign (turquoise, positive value, gray negative values), significance levels were indicated. The model fit splits patients into two patient groups (broken line). Sample number: nP = 45. B Association of underlying diseases with LMM patient groups. Cystic fibrosis (CF), chronic obstructive pulmonary disease (COPD), pulmonary fibrosis (Fibrosis), idiopathic pulmonary fibrosis (IPF). Association strength was color coded, size depicts normalized strengths, χ2 =9.5, p=0.02. Patient number: nS = 19. C Differential abundance of patient-associated viruses in LMM patient groups. Abundances were 4th root transformed. Wilcoxon, p < 0.0001. Sample number: nP =30