To my understanding, all price data is by default dividend/split normalized. However what happens if I want to filter by stock price?

Let's for example take a stock that was once priced at $10 on 2010 and doubled its price each year to be $20 in 2011, $40 in 2012, $80 in 2013. in 2014, after reaching $160, they decided to split it 4:1 so it again costs $40.

now let's say my algorithm filters out from the universe stocks that cost less than $5. does it mean that in 2010 it wouldn't buy the stock (which now has an adjusted price of $2.5?

If so - is there a way to negate it? because after all, at the original price of $10 I'd want to buy the share but on 2.5 i wouldn't due to my commission fee structure and other reasons. But o the other hand I don't want my algorithm results to get screwed up by data that is not adjusted. So what do you do in cases like that?