I read Biblical Greek. John chapter 1 reads the same in virtually all master texts, except for variants in verse 18. The questionable renderings in the New World Translation are not based on whether one is using, say, the UBS as most modern scholars do versus, say, the Westcott and Hort which the Witnesses employed, at least not here.
The reading of John 1:1, according to the Witnesses, should be "a god" at the end because, as they state, the last use of THEOS in this verse is anarthrous (occurring without the definite article). Without the definite article, the Witnesses claim, a noun becomes a predicate in relation to the verb. "In Koine Greek," goes the JW argument, "all nouns have a definite article."
This is just a lie.
The word THEOS aappears throughout John without the definite article and it never gets rendered as an indefinite noun in the NWT regardless of its position to the verbs around it.
The word THEOS is not only anarthrous at John 1:1 but also at 1:12, 1:13, and twice in verse 18. The NWT renders THEOS with an "a" in 1:1 but does not in these other places.
The reason the author did not place an article before these and other occurrences of THEOS is simple: ancient languages didn't strictly follow rules like modern ones. There is no pattern as to why the author of John employs an article for THEOS in some instances but avoids it, especially in the first chapter.
Jehovah's Witnesses are simply using people's lack of knowledge of Greek to fool others. You can even use an interlinear to check this out. There is no rule being followed by the author. They just weren't sticklers for the rules back then. It is Koine Greek 101, to be honest. You don't even have to be or rely on a scholar to see it for yourself.
One thing is clear, however, that the THEOS at the end of John 1:1 is the same as the one in 12, 13, and twice in 18. Since the others aren't rendered "a god" then neither should John 1:1 be treated any differently.