Unique reagent identifiers

The output from the enumeration should be a virtual library containing information about the molecular structure and a unique identifier as a minimum. Other useful information includes the reagents used to make a particular library member, and possibly their molecular weights (this can be used later in the purification of the synthesised library, to determine the level of impurities). 

The composition and conditions of use for ten selective staining reagents, when used in conjunction with two-dimensional solvent systems (H9), permits the unambiguous identification of seventy-six compounds of biochemical interest. Twenty-seven of these compounds may be identified specifically by a single reagent which is capable of producing a unique color (or fluorescence) sequence when the spot is viewed under successive conditions. 

Library production starts with the addition of the first position of diversity (monomer A). Four beads from each of the 384 wells are used for the QC assessment to ensure homogeneity in the coupUng reaction across all beads in a given well and identify any automated synthesizer errors in the delivery of reagents. In cases where specific monomers do not couple properly to the resin or fail the QC analysis, these monomers are eliminated from the rest of the library. After completion of the second step (monomer B), there is maximally 384 unique compounds spatially and mass encoded in the reaction plates. Again, four beads from each well are used in the QC process. 

Clearly the question of whether rearrangements do actually take place in partial hydrolysis experiments can only be solved by experience. If any such reactions occur it would be impossible to interpret results in terms of a unique sequence of amino acids unless the syntheses are completely specific and quantitative, which is most unlikely. In fact, using concentrated acid at low temperatures it has been possible to work out a unique sequence for gramicidin S (Consden et al., 1947b from the peptides identified, and there was no evidence of any peptide that did not fit this sequence. Similarly a unique structure could be determined for the phenylalanyl chains of insulin (p. 54). It thus seems unlikely that any rearrangement occurs under the action of this type of reagent. On the other hand, syntheses have been definitely shown to occur in the presence of proteolytic enzymes. The formation of plastein by the action of pepsin or trypsin on concentrated peptide mixtures has clearly been shown in certain cases to be accompanied by a decrease in amino... 

---