Info

20,000

15,000

lo.ooo

5,000

0.0-0,1 0.1 - 0.2 0.2-0,3 0.3-0,4 Deviation Range, m'z

Fig. 4.2 Serum ASMS screen results. Statistics are shown for six primary ASMS screens run against various dilutions of fetal calf serum. Compounds picked, number of compounds corresponding to ligand peaks within a defined m/z range of the center of the ligand peak. (A) Complete statistics for both positive and negative ion mass spectra; all compounds within a 0.5 m/z unit range of ligand peaks are included. Unique compounds are obtained by combining experimental data and removing duplicates. MurF Hits/Serum Hits: percentage of the 1147 matched unique compounds described in the text that intersect with the unique compounds identified from specific experiments 1-6 or that intersect with the combined list of unique compounds. Serum Hits/Total Library: number of unique compounds identified in specific experiments 1 -6 (or the combined list of unique compounds) divided into the total library of 123 405 compounds. In the last line of the table ''Total Unique'', the results from experiments 1-6 are summarized. The first three entries represent the total number of different entries from experiments 1-6 in each column. The ratios given in last two entries are calculated from the total number of unique serum hits (i.e., 36 748). (B) The expected m/z positions of compounds mostly occur close to the center of ligand binding peaks. Included from [10] with permission from SAGE Publications.

against prepared serum diluted to 2%, 10%, and 20% of its neat concentration (the latter corresponding to approximately 0.1 mM albumin, thus providing the desired low stringency). All compounds whose exact monoisotopic mass falls within 0.5 Da of a peak were annotated.

Fig. 4.2A shows the percentage of compounds on the MurF primary hit list that overlap with the primary hit list for each serum screen and the percentage of the entire library within the serum hit list. Note that these range from 44% to 50%, so that in each case the serum screen is hitting MurF ligands with higher than random (13-15%) frequency. If MurF bound only selective ligands then the serum list would be irrelevant, and one would expect the percentage of MurF hits on the serum list to be no higher than that determined by random chance based on the percentage of the entire library contained on the serum hit list. The decision to use the combined list of 36 748 compounds that occur at least once in any of the serum screens as the PCF list was made on the basis of efficiency. The highest percentage overlap with serum compounds occurs when the MurF hit list is compared to the combined PCF list. In this case, 65% of the MurF hits are from the PCF list, even though it only contains 30% of the total library. Since the majority of the unique compounds within 0.5 m/z units of a peak are very close to the center of the peak, as indicated by a graph of the distribution of deviations (Fig. 4.2B), we included all compounds within 0.5 m/z units to further ensure that the PCF list contained all possible serum protein ligands.

Was this article helpful?

0 0

Post a comment