Not to put a damper on your project, but this project is quite an undertaking, and there is a lot to consider. I would perhaps work on small bits of this, and maybe release it in small sections. This will give you the time to do it right by solving any details you need to in the smaller modules. The other issue to is that this route will be huge! The DLS will reject it if it's too big, and then you'll need to host the route elsewhere, or contact Auran if you want to upload it. Remember the mileage between Boston and Washington is about 400 miles. There are 200 miles along, give a take a few, between NYC and Washington DC alone.
How accurate is your terrain going to be?
If you are using DEM files for the line, then there will be lots of cutting to do in order to remove unnecessary baseboards. There is also the issue with the accuracy of the rail grade and the terrain.
How much outside content is there going to be?
The NEC, which I'm sure you know, is filled with lots of buildings, factories, cities, railyards, catenary, roads, highways, etc. There needs to be the balance of what looks good and speed optimization. It doesn't do any good to have a nice looking route, if it can't be driven by anyone because the machine requirements are that of a grid network made up of CRAY computers.
And related to the question above. How much custom content are you willing to create? Each region that the NEC travels through has its own unique buildings. The buildings in Phildelphia and Baltimore are much different than those in Providence and Boston. There are also lots of unique buildings, bridges and tunnels that exist no where else in the world. These too will have to be created.
A route this long can become boring. People will lose interest as they drive the long distance between Washington and Boston. The scenery will become boring, the distance between stations will seem even longer, particularly as the driver goes from NY to Boston.
There are a lot more things to think about, and consider this as a starting point. Realism is wonderful, and works in small areas, but when creating a huge route, such as the complete NEC, many compromises will need to be taken for both operational interest and computer speed.
John