Even if you could record it properly, without significant background noise (e.g. in an anechoic chamber), you'd likely not be satisfied with the results.
The actual recorded sound of rubbing a thread will sound quite different from what one would expect it to sound like. People won't recognize it.
The foley department in any movie studio is filled with objects and devices for making sound effects to be dubbed into the film's sound track.
But the sounds they make are what the audience expects something to sound like, not what it actually sounds like. (E.g. the snick-snick sound of a semi-automatic gun or rifle is usually made with a metal doorknob and lock-set, not with an actual weapon.)
I once was asked to record some sound effects for a radio play, so I naïvely took the tape recorder and recorded the requested sounds. One effect turned out particularly bad: bricks falling off a roof (and narrowly missing the hero).
I had taken a couple dozen bricks, balanced them on a wall, and pushed them off.
It sounded amazingly great at the time, but when played back in the studio it sounded like dishes being broken in a china shop (in fact, the tape was later used for exactly that purpose). Instead the director improvised and had the actors throw a few cardboard boxes on the floor while they were making the voice recording. When played back, it sounded exactly like what one would think falling bricks would sound like (heavy thudding with lots of bass).
For a thread sound, perhaps try rubbing a wooden stick along a guitar string.