1. Selection of Oligonucleotide Probes for Protein Coding Sequences
Xiaowei Wang and Brian Seed
Bioinformatics, 19(7), 2003

2. (C) 2003, SNU Biointelligence Lab, http://bi.snu.ac.kr/
Introduction
Most oligo collections
Contain poorly validated sequences
Biased toward untranslated regions(3’ UTRs)
Presumption that oligo dT will be used to prime the RNA populations
Sequence divergence is typically greater in the regions
A strategy for picking oligos for microarrays that focus on design universe consisting of protein coding regions
(C) 2003, SNU Biointelligence Lab, 

3. (C) 2003, SNU Biointelligence Lab, http://bi.snu.ac.kr/
Process
Probe candidates creation from target sequences
Filtering by oligo picking criteria
Representative probe selection
ACGCGTCGCGAGGCCTAGGCC…
Probe candidates
(C) 2003, SNU Biointelligence Lab, 

4. (1) Location in the sequence
Random priming
Result in the creation of probes that have significant representation of non-coding RNAs
Oligo dT priming
Result in probes that are enriched for mRNAs, but that will not necessarily encompass the entire coding region
Use random primed labeling because we interested in understanding the contributions of different coding region isoforms.
(C) 2003, SNU Biointelligence Lab, 

5. (C) 2003, SNU Biointelligence Lab, http://bi.snu.ac.kr/
(2) Tm uniformity
Melting temperatures should fall in a narrow range
Evaluate all sequences in the data set
Determine the median Tm for oligos by
Probe candidate is discarded if its Tm is not within 5℃ of the median Tm
(C) 2003, SNU Biointelligence Lab, 

6. (3) Probe accessibility
Consider secondary structure
The likelihood of secondary structure is greatest in regions of significant self-complementarity, for example in the stems of stem-loop structures
Tested for homology to the complementary strand of their cognate sequences using BLAST
(C) 2003, SNU Biointelligence Lab, 

7. (4) Reduced cross-hybridization
Single mismatch can be expected to destabilize the hybridization complex.
Rejection of contiguous sequence identity
Quick search using hash table data structure
(C) 2003, SNU Biointelligence Lab, 

8. (C) 2003, SNU Biointelligence Lab, http://bi.snu.ac.kr/

9. (5) Evasion of non-coding RNA and low complexity regions
Sequence regions similar to rRNA or snRNA are avoided during the probe selection process by using both contiguous base match screening and BLAST
Low-complexity regions are likely to contribute to cross-hybridization (identified by DUST program(Hancock and Armstrong, 1994))
(C) 2003, SNU Biointelligence Lab, 

10. (C) 2003, SNU Biointelligence Lab, http://bi.snu.ac.kr/
Results
60,000 mouse sequences in NCBI protein database
70mer probes
Tm: 79℃±5℃
(C) 2003, SNU Biointelligence Lab, 

11. Impact of oligo probe length
(C) 2003, SNU Biointelligence Lab, 

12. Impact of BLAST score filter
(C) 2003, SNU Biointelligence Lab, 

13. Impact of probe self-annealing filter
(C) 2003, SNU Biointelligence Lab, 

14. Impact of combining all screening filters
(C) 2003, SNU Biointelligence Lab,