This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: is 100 . 8 = 80. Similarly, there are 900 true negatives, so the expected number of false positives obtained by the test is 900 (1-. 95) = 45. The expected lenght of the list is therefore 80 + 45 = 125. Problem 3. (20 points) (a) Array 24 contains a percentage of present calls around 30%, clearly smaller than the value 40% typically observed in Aymetrix human arrays. This may indicate that array 24 is unreliable. 1 (b) Normalization needs to be applied to bring the average M values close to zero over all values of A. Normalization must be nonlinear (e.g. Lowess), as linear normalization would not work uniformly well over all values of A. Problem 4. (30 points) Write TRUE or FALSE for each of the following statements. (a) FALSE. (b) TRUE. (c) FALSE. (d) FALSE. (e) TRUE. (f) TRUE. 2...
View Full Document
- Fall '10
- Signal Processing