Proteomic profiles were characterized using the SOMAscan Assay (SomaLogic, Inc.; Boulder, CO, USA) at the Trans-NIH Center for Human Immunology, Autoimmunity, and Inflammation (CHI), National Institutes of Health (Bethesda, MD, USA). The basis of SOMAscan is built on the use of a new generation of protein-capture Slow Offrate Modified Aptamer (SOMAmer) reagents14 (link). Using these reagents, the SOMAscan Assay is able to comparatively evaluate protein abundance in 50 μl of serum, plasma, or other biological matrices. Generated by a technique referred to as Selected Evolution of Ligands by Exponential Enrichment (SELEX), the current 1.3 k Assay consists of 1,305 SOMAmer reagents selected against a variety of human proteins (47 % secreted proteins, 28 % extracellular domains, 25 % intracellular proteins) that belong to broad biological subgroups including receptors, kinases, cytokines, proteases, growth factors, protease inhibitors, hormones, and structural proteins.
For serum and plasma samples, SOMAmer reagents are binned into three separate groups according to the expected endogenous abundance of each SOMAmer’s cognate protein in typical human samples. Each SOMAmer reagent exists in only one of the three groupings. Serum and plasma samples (including controls) are then diluted into three concentrations (0.005%, 1%, and 40%) in order to create separate groups for high, medium, and low abundance proteins, respectively. Through this separation, the SOMAscan assay is able to quantify proteins across a dynamic range spanning more than 8 orders of magnitude. The diluted samples are then incubated with the dilution-specific SOMAmers.
Runs in the current 1.3 k Assay were performed semi-automatically with a Tecan Freedom Evo 200 High Throughput System (HTS), which utilizes 96— well plates. In this work, we also present serum runs performed manually with the former 1.1 k Assay using 32— well plates. Supplementary Figure1(a,b) shows Venn diagram comparisons between the two assays based on aptamer sequence (“SeqId”) and target analyte, respectively. For a total of 1,061 SOMAmers, the aptamer sequence remained unchanged. However, for 60 SOMAmer targets, the aptamer was replaced to improve binding affinity and specificity. The case study presented below in the Results Section presents 1.1 k Assay data from one serum run performed by SomaLogic with a Beckman BioMek Fx HTS. The total number of samples analyzed is 2,624.
The typical SOMAscan plate design includes buffer wells (no sample added), quality control (QC_SOMAscan) and calibrator samples provided by SomaLogic. Quality control and calibrators are pooled samples composed of the same matrix as the biological samples being measured in the plate. Usually, 2 sets of quality control samples are run in duplicate in each plate. Since the intended purpose of these samples is to assess the quality of measurements obtained from one single plate, quality control samples may vary from plate to plate. Also, 5 to 7 replicate calibrator samples are included in each plate with the purpose of normalization across plates. The calibrator consists of a common pooled sample used across a large number of runs; however, when that calibrator lot is depleted, SomaLogic must switch to a different calibrator lot. In addition to these, we have added bridge samples (QC_CHI) to our plate designs in serum and plasma, which we typically run in quadruplicate in each plate and keep consistent across all runs in the 1.3k Assay. Our serum QC_CHI consists of 17 (8 male, 9 female) pooled samples from healthy donors of median age 35 (Q1 = 28.5, Q3 = 54.5) years. For our plasma QC_CHI, we pooled 21 (10 male, 11 female) samples from healthy donors of median age 57 (Q1 = 37, Q3 = 61.5) years. Table1 presents summary statistics of all serum and plasma runs performed at CHI between January 2015 and April 2017, all of which were analyzed in this paper. The table includes, in parentheses, the breakdown in terms of (nrep × npl), where nrep is the number of replicates per plate and npl is the number of plates.
For serum and plasma samples, SOMAmer reagents are binned into three separate groups according to the expected endogenous abundance of each SOMAmer’s cognate protein in typical human samples. Each SOMAmer reagent exists in only one of the three groupings. Serum and plasma samples (including controls) are then diluted into three concentrations (0.005%, 1%, and 40%) in order to create separate groups for high, medium, and low abundance proteins, respectively. Through this separation, the SOMAscan assay is able to quantify proteins across a dynamic range spanning more than 8 orders of magnitude. The diluted samples are then incubated with the dilution-specific SOMAmers.
Runs in the current 1.3 k Assay were performed semi-automatically with a Tecan Freedom Evo 200 High Throughput System (HTS), which utilizes 96— well plates. In this work, we also present serum runs performed manually with the former 1.1 k Assay using 32— well plates. Supplementary Figure
The typical SOMAscan plate design includes buffer wells (no sample added), quality control (QC_SOMAscan) and calibrator samples provided by SomaLogic. Quality control and calibrators are pooled samples composed of the same matrix as the biological samples being measured in the plate. Usually, 2 sets of quality control samples are run in duplicate in each plate. Since the intended purpose of these samples is to assess the quality of measurements obtained from one single plate, quality control samples may vary from plate to plate. Also, 5 to 7 replicate calibrator samples are included in each plate with the purpose of normalization across plates. The calibrator consists of a common pooled sample used across a large number of runs; however, when that calibrator lot is depleted, SomaLogic must switch to a different calibrator lot. In addition to these, we have added bridge samples (QC_CHI) to our plate designs in serum and plasma, which we typically run in quadruplicate in each plate and keep consistent across all runs in the 1.3k Assay. Our serum QC_CHI consists of 17 (8 male, 9 female) pooled samples from healthy donors of median age 35 (Q1 = 28.5, Q3 = 54.5) years. For our plasma QC_CHI, we pooled 21 (10 male, 11 female) samples from healthy donors of median age 57 (Q1 = 37, Q3 = 61.5) years. Table
Summary statistics of all runs analyzed in this paper.
| Serum 1.3k (HTS) | Plasma 1.3 k (HTS) | Serum 1.1 k (Manual) | |
|---|---|---|---|
| Plates | 15 | 8 | 11 |
| Buffer | 19 (1 × 13, 3 × 2) | 10 (1 × 7, 3 × 1) | 11 (1 × 11) |
| Calibrators | 101 (7 × 13, 5 × 2) | 54 (7 × 7, 5 × 1) | 55 (5 × 11) |
| QC_SOMAscan | 57 (2 × 27, 3 × 1) | 31 (2 × 14, 3 × 1) | 22 (2 × 11) |
| QC_CHI | 59 (4 × 14, 3 × 1) | 28 (4 × 7) | 0 |
Free full text: Click here
