Only 73 of 100 correspondents have investigated (ie, sequenced) false positives.
Of these, the most common :
| proteins | found as false positives in IT | believed to date to be real interactors (found in IT) | found as false positives in systems other than IT # |
| hsp's | 16 | 5 | - |
| ribosomal proteins | 14 | 1 | 3 |
| cytochrome oxidase | 5 | - | 1 |
| mitochondrial proteins* | 3 | 1 | 2 |
| proteasome subunits | 4 | 3 | - |
| ferritin | 4 | - | 2 |
| tRNA synthase | 3 | - | 1 |
| collagen-related proteins | 3 | - | - |
| Zn finger proteins | 3 | 4 | 2 |
| vimentin | 2 | - | - |
| inorganic pyrophosph. | 2 | - | - |
| PCNA | 2 | - | - |
-------
* 5 other mitoch. proteins are among other categories (hsp, ribos., cyt.C oxidase, ATP-synth.)
# 19 responses analysed to date
Some investigators (data not in the database because of confidentiality
restrictions) also note for the most common trash:
| elongation factors | ferritin | ||
| lamin | cytoskeletal proteins | ||
| still more ribosomal RNA clones | |||
| ubiquitin | |||
| (#1) | (#2) |
Analyzing the few answers we received from the users of other two-hybrid
systems, we also found ribosomal proteins, ferritin and ubiquitin mentioned
prominently. Some entries on our compiled lists are easily explained based
on their biological properties (ie, hsps). Others are more mysterious -
ie, PCNA. We would be interested to hear comments/theories from interested
parties.