Genetic determinants of HIV-1 subtype C Nef-mediated SERINC3 down-regulation

Table 3 Quantile regression models showing the relationship between each Nef function and E values

Nef function ^a	Univariable analysis		Multivariable analysis
Nef function ^a	Coefficient	P value	Coefficient	P value
dE0^b
SERINC3 down-regulation	− 4.041	0.0001	− 3.548	0.002
SERINC5 down-regulation	− 2.229	0.038	− 0.835	0.464
HLA-I downregulation	− 1.647	0.145	− 1.920	0.046
CD4 downregulation	− 2.225	0.046
Alteration of TCR signalling	0 .458	0.691
dE90^b
SERINC3 downregulation	− 6.094	0.0001	− 4.678	0.009
SERINC5 downregulation	− 4.883	0.003	− 2.786	0.123
HLA-I downregulation	− 2.096	0.190	− 0.701	0.641
CD4 downregulation	− 3.634	0.020
Alteration of TCR signalling	− 0.425	0.798

^aEach Nef function was standardised by subtracting the mean and dividing by the standard deviation to calculate a z-score
^bThe dE0 and dE90 values assigned to each Nef clone were derived from the Ising and Potts models respectively and represent Nef fitness in vivo [16]. The Ising model distinguishes between consensus and non-consensus residues at each codon. The Potts model accounts for the diversity of amino acids present at each codon

ISSN: 1743-422X