Qualifiers, data base

As we have been adding to the data base at Du Pont (to the one and two group tables), the question has often been raised how much detail should we go into in the qualifiers This is somewhat of a dilemma, many of the industrial reactions we are dealing with have only been considered for a limited number of substrates. It is not clear whether qualifiers should be incorporated that restrict the transforms to only those cases where it is known to work or whether only those which are known to fail should be specifically excluded. Both ways take an immense amount of literature work to do consistently. 

We all know that butadiene can be dimerized under catalytic conditions to a wealth of different products. Addition of alkyl substituents changes the product mix and introduces a variety of different stereoisomers as well. What happens if we put functional groups on butadiene. Do all the reactions still proceed - do you get any new ones, etc. We do not know and serious doubt whether many experiments have ever been run 

ACS Symposium Series American Chemical Society Washington, DC, 1977. 

In summary, adding simple reactions to LHASA is simple. Incorporating sophisticated reactions can be as complicated as you wish to make it. (Work is currently underway to prepare a general package of subgoal transforms which will serve to remove interferences - relieving the chemist from having to work them out separately for each super transform.) 

Why is Du Pont interested in LHASA The program was clearly designed for carbocyclic natural products synthesis in mind - an area in which the Company has only limited interest. 

---

In some cases it is completely impossible to prepare control samples for the nse in X-charts. So we have only the second component from the last example. The long-term component that takes into acconnt changes between series has to be estimated differently. If we don t have other data, the last possibihty is a qualified guess based on the experience of the analyst. In the table an example is shown... 

Certain medical use qualifiers were coded into the data base. These medical use qualifiers form a fourth cut through the hyperspace, which again offers a statistically significant way ot distinguishing between kinds ot medical uses. As was seen from the distribution function tor uses/compound, the X qualifier (null character-no additional information showed a mean uses/compound which is not significantly different from the entire data base. This was also true or the z quaiitier (additional information . However,... 

The amount of rejected and qualified data varied greatly with a large increase in the fourth quarter. Samples from the first three quarters were analyzed by one laboratory, and the samples from the fourth quarter were analyzed by a different laboratory. Based upon the above results and experience at other sites, not ill laboratories perform equally. Even though both laboratories were using identical methods, quality control procedures, and statements of work designed to obtain consistent results, the resultant quality of work varied greatly. Apparently, the procedures and associated quality control were not adequately transferred to the laboratory bench. 

The MRS pumps and supports have been seismically qualified by means of an engineering walkdown inspection, and comparison with a seismic experience data base (Reference 71). 

Use Combinations. Each compound in the data base has one or more distinct use names associated with it. For some compounds, the set of names may contain two that are the same but whose (letter) qualifiers are different. These differently qualified use names will be considered as different use names for the purpose of this Investigation. Thus, each compound can be assigned an Integer which denotes how many distinct use tokens are associated with It. This Integer, which measures uses per compound, can also be assigned to each use token associated with the given compound. In the data base, the number of uses per compound varies generally up to 6, but there are a special set of compounds which have 9 associated uses each. 

Plotted in Figure 2 are the cumulative type frequency distributions for two subsets of the data base. The single use curve corresponds to all compounds with a single use. The X curve corresponds to all uses with the X (null) qualifier l.e., all uses which are unqualified. These subsets are also seen to... 

Figure 2. Log-normality of data base subsets. Points are plotted on a log, frequency scale curve represents a normal, rather than log-normal, distribution in frequency. Single uses, X uses (no qualifier), O. Norrrud curve vs. frequency fit to single-use data.

Thus X and Z compounds seem typical of the average compound in the data base, while a compound with an H or F qualifier has significantly more uses than average. This result suggests... 

We developed a search program which provides a very general search capability. It allows us to qualify the search or any piece (or pieces) of information in the data base and has considerable Boolean logic capabilities. For example, if we were Interested In compounds within the range RH-60000 to RH-80000 which were active against fungus ABC or DBF, but did not injure crop XYZ at a rate of 4 Ibs/acre, we would encode the question as ... 

A large percentage of eddy-current inspections are conducted in the field, away from the home base and often in remote or inaccessible locations. Using local telephone lines or mobile phone lines would allow the inspector to beam his data back to the office. In this way highly qualified personnel can be consulted when problems or difficult to interpret results occur. Inspectors no longer need to feel isolated on site. 

To obtain some idea as to the actual vertical penetration of TCDD in those areas contaminated by dispersion, three sample points were selected near the farm road which was sprayed in 1971. At each point a single sample was collected from each of three depths, viz, at the 0- to 2-, 2- to 4-, and 4- to 6-inch depths, using clean knives and spoons. The selection of a 2-inch increment was based upon the estimated removal on a single pass by a qualified operator of earth-moving machinery. The resulting data are as follows ... 

This chapter deals with the necessity of representative sampling in the context of PAT. All PAT sensors need to be calibrated with respect to relevant, reliable reference data (Y data). This presupposes that representative samples are at hand for this characterization - but sampled howl Additionally, X signals (X measurements) need to be qualified as representative of the same volume as was extracted for Y characterization, or at least a sufficiently well-matching volume. How does one demonstrate this in a quantitative manner If the quality of both X and Y data involved is suspect, how can a multivariate calibration be expected to be trustworthy This also includes the issue regarding proper validation of the chemometric multivariate calibration(s) involved, which can only be resolved based on proper understanding of the phenomenon of heterogeneity. The TOS delivers answers to all these issues. The TOS constitutes the missing link in PAT. 

An official O M program is to be developed based on the inspection and assessment data, as soon as possible if ACM is located. Either the Asbestos Program Manager or a qualified consultant should develop the O M program. The written O M program should state clearly the O M pohcies and procedures for that building, identify and describe the administrative line of authority for that building, and should clearly define the responsibilities of key participants,... 

---