|
Activity Number:
|
154
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #310051 |
|
Title:
|
Outlier Detection for a High-Throughput Bioassay
|
|
Author(s):
|
Shea Watrin*+ and Ye Frank
|
|
Companies:
|
Amgen Inc. and Amgen Inc.
|
|
Address:
|
4000 Nelson Rd, Longmont, CO, 80503,
|
|
Keywords:
|
Bioassay ; Outliers
|
|
Abstract:
|
Bioassays are used to evaluate the potency of pharmaceuticals. These cell based assays can produce extreme results for which an assignable cause is not apparent. In a high throughput lab environment it is desirable to quickly and efficiently discern outliers from good data points. Investigating individual outliers is time consuming and can lead to inconsistent rejection of data. It is therefore desirable to set rules that will allow rejection of outliers automatically. These rules should protect against outliers but allow for variability that is common in the assay. We establish guidelines that allow the data to be evaluated for outlying data points before assumptions of the assay are verified and calculation of potency is carried out. These guidelines protect against subjective decisions about outliers and allow a high-throughput lab to process data efficiently.
|
