The number of target peptides is not the same as the number of decoy peptides

[zrolfs]

This appears to be an N-terminal Methionine related issue. It is observed independent of variable modifications, but disappears without an M on the N-terminus. I suspect there are two problems: one with an N-terminal duplication and another with variable modifications generating combinations (Example: MARM -> MAR, M(Ox)AR, M, and M(Ox) [4] vs the decoy MMRA -> MMR, M(Ox)MR, MM(Ox)R, M(Ox)M(Ox)R, and A [5]).

Here's a unit test that fails:
[Test]
public static void DecoyIsBroken()
{
Protein targetProtein = new Protein("MPRTEINSEQENCEWITHAKANDANRANDSMSTFF", "accession1");
var decoyProtein = DecoyProteinGenerator.GenerateDecoys(new List { targetProtein }, DecoyType.Reverse)[0];
DigestionParams dp = new DigestionParams();
var targetPeptides = targetProtein.Digest(dp, new List(), new List()).ToList();
var decoyPeptides = decoyProtein.Digest(dp, new List(), new List()).ToList();
Assert.AreEqual(targetPeptides.Count, decoyPeptides.Count);
}