Figure 1. Plasma protein concentrations. Graph showing a proposed log–log relationship between plasma protein concentration and number of proteins at the various concentrations. Some well-known proteins are indicated at their normal plasma concentrations showing a known dynamic range spanning 12 orders of magnitude (albumin to tumor necrosis factor). The advantage of large sample size is indicated by the difference between the dotted line (1 nm, sensitivity 100 fm, sample size 0.1 ml) and the dashed line (0.1 pm, sensitivity 100 fm, sample size 1 l). (Original slide courtesy of D. Hochstrasser, modified.) mm: millimolar; m: micromolar; nm: nanomolar; pm: picomolar; fm: femtomolar; am: attomolar; zm: zeptomolar; ym: yactomolar.

Figure 2. In-depth industrial-scale proteomics. A portion of each of a set of carefully selected samples from patients with a disease is pooled, while separate portions are retained unpooled. Small squares represent an example of a protein common to all diseased samples, and the small triangle represents a protein present in one particular sample (phenotypic variation), prior to and after extensive separation. Identified/characterized proteins populate a database. Similar analysis of a control set permits comparisons to be made. The unpooled portions of the individual samples are used for screening for proteins of interest using arrays or other appropriate high-throughput assays.

Figure 3. Donor sample-based high-throughput proteomics. A large set of small samples (here 10000) from patients with a disease is analyzed. Similar analysis of a control set permits comparisons to be made.