[Reproducibility of protein profiling update]
I'm rereading the literature to see if I can find examples of the various trained algorithms successfully classifying blind sample sets that are new to them. Anyone finds one, please post. This one is not encouraging.
>>Cancer Research 63, 6971-6983, October 15, 2003]
© 2003 American Association for Cancer Research
Proteomic Profiling of Urinary Proteins in Renal Cancer by Surface Enhanced Laser Desorption Ionization and Neural-Network Analysis
Identification of Key Issues Affecting Potential Clinical Utility1
Mark A. Rogers, Paul Clarke, Jason Noble, Nicholas P. Munro, Alan Paul, Peter J. Selby and Rosamonde E. Banks2
Cancer Research UK Clinical Centre [M. A. R., P. C., N. P. M., P. J. S., R. E. B.] and Department of Urology [A. P.], St. James’s University Hospital, Leeds LS9 7TF, and School of Computing, University of Leeds, Leeds LS2 9JT [J. N.], United Kingdom
Recent advances in proteomic profiling technologies, such as surface enhanced laser desorption ionization mass spectrometry, have allowed preliminary profiling and identification of tumor markers in biological fluids in several cancer types and establishment of clinically useful diagnostic computational models. There are currently no routinely used circulating tumor markers for renal cancer, which is often detected incidentally and is frequently advanced at the time of presentation with over half of patients having local or distant tumor spread. We have investigated the clinical utility of surface enhanced laser desorption ionization profiling of urine samples in conjunction with neural-network analysis to either detect renal cancer or to identify proteins of potential use as markers, using samples from a total of 218 individuals, and examined critical technical factors affecting the potential utility of this approach.
Samples from patients before undergoing nephrectomy for clear cell renal cell carcinoma (RCC; n = 48), normal volunteers (n = 38), and outpatients attending with benign diseases of the urogenital tract (n = 20) were used to successfully train neural-network models based on either presence/absence of peaks or peak intensity values, resulting in sensitivity and specificity values of 98.3–100%. Using an initial "blind" group of samples from 12 patients with RCC, 11 healthy controls, and 9 patients with benign diseases to test the models, sensitivities and specificities of 81.8–83.3% were achieved. The robustness of the approach was subsequently evaluated with a group of 80 samples analyzed "blind" 10 months later, (36 patients with RCC, 31 healthy volunteers, and 13 patients with benign urological conditions). However, sensitivities and specificities declined markedly, ranging from 41.0% to 76.6%. Possible contributing factors including sample stability, changing laser performance, and chip variability were examined, which may be important for the long-term robustness of such approaches, and this study highlights the need for rigorous evaluation of such factors in future studies. <<
And here's a full text freebie that deals with the issue:
biomedcentral.com
Cheers, Tuck |
