Generating random start and target nodes

Hi there,
I hope some of you can help with an issue I an having .
I want to select a random starting node with maps to a random target node through a single specific relationship. I tried the following query but the target node is not random and when I add a distinct function to the target node it return and error.
MATCH (a)-[r:causes_f5plus]->(t:dsyn)
RETURN distinct(, type(r),, round(rand(), 10) as score
order by score
limit 50
/// this return random starting nodes but non-random target nodes.

I tried this and it return an error.
MATCH (n)-[r:causes_f5plus]->(m:dsyn)
RETURN distinct(, type(r), distinct(, round(rand(), 10) as score
order by score
limit 5
// but it return ..
Unknown function 'distinct' (line 2, column 35 (offset: 73))
"RETURN DISTINCT (n).name AS (, type((r)) AS type(r), distinct(((m).name)) AS distinct(, round((rand()), (10)) AS score

I also tried collect(distinct( and it is not random.

I wish some of you can lend some help.

by the way when I say non-random results I mean something like this
I am looking from something like this where both the starting nodes and the target nodes are completely random.
your help is greatly appreciated.

Can you please apply Distinct keyword in match rather than in return statement.
Thanking you
Yours faithfully
G Sudhir Sameer

thank you very much, it seems to be working.