Drug targets are often transmembrane proteins; however, we lack an understanding of how these receptors assemble. Alpha7 nicotinic receptors (a7R) require specific chaperone proteins to assemble in human embryonic kidney (HEK) cells, while related 5HT3 serotonin receptors (5HT3R) do not. Resistance to Inhibitors of Cholinesterase 3 (RIC3) and Transmembrane Protein35a (TMEM35) are two established chaperones for a7Rs. Chimeric a7R -5HT3 receptors substituting 5HT3R for a7R transmembrane domains (TMs) express well in HEK cells without chaperones. However, none of the chaperones allowed the assembly of chimeras with mixed a7R and 5HT3R TMs, suggesting TM amino acid side chains must be compatible. Many amino acid side chains for TM4 extend into membrane lipids and do not interact with side chains from TM1-3. We hypothesized substituting a7R side chains for those 5HT3R TM4 side chains projecting into lipids is inconsequential. However, the TM structure of a7R is unknown, so we modeled a7R TM4 based on homology with 5HT3R and other receptors using YASARA, UCSF Chimera/Modeler, and USCF Chimera/Matchmaker. From these models, we made two sets of mutations, conservative substitutions (CS) in which only side chains farthest away from TM1-3 were substituted and less conservative substitutions (LCS) changing all side chains except for those interacting with TM1-3. In the preliminary experiment (N=1), all constructs expressed in HEK cells, CS, and LCS showed reduced but significant expression. We will test whether RIC3 or TMEM35 can increase expression; preliminary data suggests our hypothesis is correct.