The goal of our past and ongoing research is to understand, explore, and evaluate the existence of previously unknown biomarker information within the lower molecular weight range of the circulating proteome (e.g. serum, plasma, nipple fluid). None of this work should be construed or represented as a clinical test, or the evaluation of a commercial clinical test.

We are currently focused on sequencing and identifying the low molecular weight biomarkers that underlie the mass spectral signatures. For this goal we are taking advantage of our recent finding that indicates that the vast majority of the low molecular weight biomarker proteins are amplified by binding to circulating carrier proteins such as albumin. We plan to post lists of differentially expressed peptide sequences for a variety of diseases. Currently we are in the process of antibody based validation for many low abundance novel markers.

Proteomic Fingerprinting

The following study sets were independently obtained using different experimental conditions, mass spectrometry platforms, different sampling handling conditions, and employ different clinical study sets. These samples come from both human and animal sources and are provided anonymized. It is widely known that MALDI-TOF spectra are highly dependant on the specific conditions used to generate the signal. Moreover, each of these study sets are different in one or more important parameters (e.g. type of chip surface, type of mass spectrometer, sample handling, source and type of body fluid) . In fact, most of our studies were generated with the expressed intent to study and evaluate the effect of the changing experimental conditions on the characteristics of the spectra. Thus, while we offer these study sets to the scientific community in an unrestricted manner, we caution against any direct comparative analysis within or between sets. Certainly, we highly recommend that you contact us before making such comparisons, and that you employ appropriate normalization, baseline correction, and alignment prior to any such analysis.

High Resolution SELDI-TOF Study Sets

1. Ovarian cancer case vs. high-risk control: This study was published in The Endocrine Related Cancer Journal in June 2004. We have attached as well an example report on the quality control/quality assurance for this set. Please click on "Ovarian cancer QA/QC report" for a Word download.

   GOALS OF STUDY: We generated this data, using a non-randomized study set of ovarian cancers and control specimens on an ABI Qstar fitted with a SELDI-TOF source to begin to collect data relative to critical unanswered questions in the field of proteomic profiling as follows:

   A. Does the use of high-resolution time-of-flight (TOF) mass spectrometry (MS) for gathering proteomic fingerprints from surface-enhanced laser/desorption ionization (SELDI) ProteinChip arrays yield better analytical and clinical sensitivity and specificity compared to low-resolution instrumentation, at least for the set of serum samples analyzed within this study? According to NCCLS experimental design criteria, a single variable was isolated (i.e., the type of mass spectrometer) to answer this question. Since we were able to analyze the exact same SELDI ProteinChip spot with two different mass spectrometers, a direct comparison could be attempted.

   B. Which contributes the greatest source of variability: the heterogeneity within and between clinical serum sets (i.e., normal study set vs. cancer study set) or the sample application/MS process itself? To address this question, reference standards were used and the variable to be tested was fixed and isolated. Thus, we could not randomize and co-mingle the cancers and controls at the same time since we wanted to determine the variability within the mass spectrometer/sample application within a given run cycle and test that variable on a common set of samples - we chose not to co-vary two independent variables at one (run date and phenotype). This experimental design allowed us to answer the important question and show that for this study set, the variability within the process itself was greater than the variability within the samples sets.

   C. Does the development of spectral QA/QC procedures positively impact the modeling performance? In other words, does elimination of “bad looking spectra” contribute to better performing models?

   Acquire High Resolution Ovarian Data (Zip Format, 275 MB)

2. Premalignant Pancreatic Cancer Detection: Published Cancer Cell, December 2003
   • Download .pdf of Cancer Cell Paper
   • Acquire High Resolution Pancreatic Cancer Premalignant Data (7.4MB Zip File)

   GOALS OF THIS STUDY:
   The goals of this study, using a randomly commingled study set of murine sera, were to explore the ability of the low molecular weight information archive to classify and discriminate premalignant pancreatic cancer compared to control animals.

3. Toxicoproteomic analysis of anthracycline-induced cardiotoxicity: Published Toxicological Pathology, March 2004
   • Download .pdf of Toxicol. Path Paper
   • Acquire High Resolution Tox Path Data (293MB Zip File)

   GOALS OF THIS STUDY:
   The goals of this study, using a randomly commingled study set of murine sera, were to explore the ability of the low molecular weight information archive to classify and discriminate animals exposed to cardiotoxic compounds vs vehicle alone. This study employed a randomized and double blinded design with a blinded validation set.

4. Wegeners Disease Study. This study is in press at Arthritis and Rheumatism. For collaborative contacts and study details please email Dr. John Stone, the Director of the The Johns Hopkins Vasculitis Center at : jstone@jhmi.edu.

   Acquire High Resolution Wegeners Data (467 MB Zip File)

   GOALS OF THIS STUDY:
   The goals of this study were to explore the ability of the low molecular weight information archive to discriminate patients with active Wegeners granulomatosis from those in remisssion using a randomly commingled study set of serum from patients. Wegeners granulomatosis, is a vasculitic inflammatory condition.

Low Resolution SELDI-TOF Datasets

1. Ovarian Cancer Studies
   1. Data from publications: Download .pdf of Lancet paper, published February 2002

      GOALS OF THIS STUDY:
      This data was collected using the H4 protein chip and a Ciphergen PBS1 SELDI-TOF mass spectrometer. The chip was prepared by hand using the recommended protocol. The spectra were exported with the baseline subtracted. The goal of this feasibility study was to explore the existence of low molecular weight information that could serve as a diagnostic classifier, using a defined study set as a test ground.

   2. Data from unpublished experimental studies
      1. 4/3/02 Ovarian Study set
         GOALS OF THIS STUDY:
         Due to the discontinuation of the H4 chip, the WCX2 chip was chosen as a replacement. The Lancet study was repeated using the WCX2 chip. These samples were processed by hand and the baseline was subtracted creating the negative intensities seen for some values.

         Ovarian Dataset 4-3-02.zip (31 MB)

      2. Ovarian Dataset 8-7-02: x
         
         GOALS OF THIS STUDY: This dataset while produced using the WCX2 protein chip, differed greatly from the 4/3/02 study set. The goal of this study was to explore the impact of robotic sample handling (washing, incubation, etc.) on the spectral quality. We employed an upgraded PBSII SELDI-TOF mass spectrometer to generate the spectra. Different sets of ovarian serum samples were used compared to previous studies. The sample set included 91 controls and 162 ovarian cancers, which were not randomized so that we could evaluate the effect of robotic automation on the spectral variance within each phenotypic group.

         Ovarian Dataset 8-7-02.zip (31 MB)

2. Prostate Cancer Studies
   Download .pdf of JNCI paper, published October 2002

   GOALS OF THIS STUDY:
   This data was collected using the H4 protein chip and a Ciphergen PBS1 SELDI-TOF mass spectrometer. The chip was prepared by hand using the recommended protocol. The spectra were exported with the baseline subtracted. The goal of this feasibility study was to explore the existence of low molecular weight information that could serve as a diagnostic classifier, using a defined study set as a test ground.

   JNCI Dataset 7-3-02.zip (46 MB) 

Other Recent Publications from our Laboratory

1. BMC Bioinformatics Rebuttal
2. OvaCheck Question and Answer
3. Carrier Protein Binding and Biomarker Amplification
4. Written in Blood- Nature Concepts
5. Current Opinion in Biotechnology
6. Cancer Diagnosis Using Proteomic Pattern
7. Proteomic Applications for the Early Detection of Cancer
8. Mass Spectrometry-based Diagnostics: The Upcoming Revolution in Disease Detection
9. SELDI-TOF Fingerprinting and Pattern Profiling of Human Cancer Tissue

We welcome any comments or suggestions about how we can improve the website. We are attempting to provide a useful and current database that will allow investigators to perform their own analysis. Please send your comments to Dr. Gordon Whiteley.