Table I Comparison of statistical methods for identifying intact proteins in the top-down fashion
False discovery rate (q value cutoff) (%)Number of unique identificationsp score cutoffE-value cutoff (simple proteoform databasea)E-value cutoff (complex proteoform databaseb)
0.103671.2 × 10−102.0 × 10−52.6 × 10−3
1.04074.7 × 10−97.7 × 10−40.10
5.04919.2 × 10−81.5 × 10−22.0
  • a Proteoforms are created in a candidate expansion approach called “shotgun annotation” (26); the simple database contains 164,088 candidate proteoforms.

  • b The complex database contains 21,624,023 candidate proteoforms (see “Experimental Procedures”).