We demonstrated that PTMClust improved on both true positives (correct modification position predictions) and false positives (misplaces modification positions) when applied to the outputs of SIMS, InsPecT, MODmap, and InsPecT post-processed with PTMFinder, a PTM refinement algorithm.
This influence is considerably less for geometric modifications positioned on the exterior of the isolator.
Further, this shows that thiophosphate siRNAs with the same nucleotide sequence but with different sulfur modification positions have different silencing effects.
Similarity, for InsPecT, our algorithm reduced the number of misplaced modification positions by ∼38%9191 fewer modification position misplacement) to produce 712 correct predictions, an increase of ∼15%.
Similarly, InsPecT result matched 860 reference sequences but misplaced 239 modification positions.
Moreover, many instances of the same modified peptide either share the same modification position (for both correct and misplaced cases) or have vastly different modification positions that point to different phosphorylation sites in the peptide.
The true modification position for each peptide was randomly chosen to be on one of the instances of the preassigned amino acid for that subset, and the modification positions used as input to the algorithms are set to a noisy version of the true modification positions.
We next examined the abilities of our algorithm to identify PTM groups corresponding to bona fide PTMs, fine-tune observed modification masses and correct for misplaced modification positions.
More specifically, for SIMS, PTMClust decreased the number of misplaced modifications by ∼40% (106 fewer misplaced modification positions) to produce 791 correct matches, an increase of ∼15%.
However, these search methods suffer from two problems: mass measurement inaccuracy and uncertainty in predicting modification positions, which limit their accuracy and precision.
Given input data with higher mass resolution and fewer misplaced modifications due to cleaner ion fragmentation signals, PTMClust can improve upon its abilities to refine modification positions and find meaningful PTM groups.
