I've always been surprised by how low the torque specs for AN bolts are compared to similar strength and size bolts in general engineering application. It is true they are predominantly used in shear applications, so it isn't very important. But some applications are tension, or combined, and deserve higher torques.
General engineering practice calls for torques that produce preloads between 60-80% of yield strength of the bolt, and as others have said, this gives the best fatigue resistance and strength and prevents joint gapping for general-purpose loads that have some varying content. Bolts holding the engine mount to the fuselage would qualify under this practice. You don't know or analyze the varying loading in detail, but want the longest service life.
People think that over torquing uses up available strength margin, but that is incorrect. A pre-loaded bolt sees (almost)* no change in bolt stress or elongation until the applied load exceeds the torque preload. For applied loads below the torque pre-load, the combination of the applied load and the clamping pressure from the parts maintains a (nearly)* constant total load in the bolt. At the point where the preload is exceeded, the bolt is now carrying only the applied load and elongates farther, causing minute gapping of the joint. Gapping and motion in the joint are undesireable. Higher preloads create stronger joints. An exception to this is in predominantly shear applications, where the tensile preload reduces slightly the shear strength (see: Mohr's circle)
If you want to compare the AN specs to general engineering practice, look up a torque table for SAE grade 5 bolts of the same thread. SAE grade 5 is 125ksi, same as AN. (for bolts 1" dia. or less) For a 3/8-24 bolt, the dry torque spec is 420 in-lbs, the lubricated-thread torque spec is 300 in-lbs. But an AN6 (3/8-24) spec is 160-190 in-lbs (dry). If you add to that the fact that we are usually torquing an elastic lock nut or a metal lock nut, which takes torque to turn, I think we often have significantly undertorqued bolts.
* the 'almost' caveat here is because the parts clamped by the bolt do compress very slightly, and as the load varies, some of the compression in the parts relaxes as the applied load supplies the tension rather than the clamping of the parts. The joint becomes a system kind of like a suspension, where the deflection in the spring (bolt) matches the deflection in the parts. The more rigid the parts being clamped, the less variation in bolt stress for loads less than the torque preload. But nothing is infinitely stiff, there is always a minute amount of compression in the clamped parts.
"The Magic Carpet"
Hobbs 500 in 8-3/4 years
also LS-6-15/18 sailplane
VAF donation Dec 2017