Biology Reference
In-Depth Information
the data. . . . When you do a standard tracking sheet, you know
it's there, and it's always there. You also enforce, or at least you
can see, that it's been done the same way over and over again.
This is a production environment and for us variability is hard
to deal with, we want to have as little variability as possible and
standard tracking sheets are very good for that.
The detail with which such tracking is performed is illustrated in the
kinds of checklists that Vokoun proposed for the MBPG. The checklist
in fi gure 3.7 shows that workers had to account for the numbers of
tips, wipes, tubes, and caps at their workstation each day. Their manag-
ers then used a sheet to score their work on the basis of “shiny clean”
fl oors, “unused pipettes, tools, [or] fi xtures” cluttering their workspace,
maintenance of checklists, and so on.
39
All this monitoring depends critically on machines. It is computers
that maintain not only the detailed monitoring data, but also the care-
ful control over space and people. “Our life is spreadsheets,” Meredith
told me simply. “We love spreadsheets, we hate spreadsheets.” But Mer-
edith also told me that some of their needs for managing data had far
outgrown the capacity of spreadsheets. By now it was really databases
that ran the lab: SAP and the Broad's laboratory information manage-
ment system (LIMS), called SQUID. In one way at least, the cephalopod
name is appropriate: SQUID's tentacles extend in all directions into all
corners of the laboratory, sucking data back to a central repository. Any
sample that passes through the lab leaves its trace in SQUID—the his-
tory of its movement is recorded in painstaking detail. Natalie, an asso-
ciate director of the Genome Sequencing Platform, described her work
coordinating and managing sequencing projects. Projects were initiated
by generating records in SQUID, projects were monitored by watching
their progress through SQUID on a daily and weekly basis, and projects
ended when they were removed from the database. At the Broad, the
production of sequence becomes an information management problem.
The ability to manage leanly, to create spaces and people amenable to
the principles of operational analysis, means having the ability to mea-
sure, to quantify, and to track.
The Broad's raw material is samples (it deals with thousands); its
products are bases of DNA sequence (it produces billions per year);
in the middle, petabytes of data are generated. Measuring, quantify-
ing, and tracking are possible only with computers. It is computers,
particularly large and sophisticated databases, that have allowed the
techniques of production management to be imported into molecular
