ChipViewer Software

1
ChipViewer Software

• ChipViewer is coded to visualize and analyze
  the tiling chip data.

• Data visualization
• Data normalization
• Data mining
• Novel gene detection and fetching
• Expression analyses

2
ABSTRACT
The SSP tiling chips, which cover the whole
Arabidopsis thaliana genomic sequences, are
customarily designed for multi-purposes To map
transcriptional units of the Arabidopsis ORFeome,
to mine and identify novel transcripts and small
RNA species, to do high-throughput polymorphism
detection among Arabidopsis accessions and to
identify evolutionarily conserved non-coding
regions in Brassica. The ChipViewer, a
Unix/Linux based application, is coded in-house
in the Salk Institute Genomic Analysis Laboratory
(SIGnAL) to qualitatively and quantitatively
analyze the huge amount of data generated bye
tiling chip experiments. Its X/Motif graphic user
interface (GUI) allows uses to view chip
expression profiles, de-convoluted chip
expression profiles and histograms. An improved
Match-only Integral Distribution algorithm
(MOID) is introduced to call gene or genomic
region absolute expression. In addition, a
Relative Weight (RW), instead of normalization,
is adopted to make it possible for universal
comparison and analysis among different chip
experiments. Furthermore, its comparative
analysis tool allows for discovery of Single
Feature Polymorphisms (SFPs) among Arabidopsis
accessions, which will be useful markers for
mapping QTLs using natural variations in the
weed. A snapshot can be reached at
http//signal.salk.edu/chipviewer/
1. ChipViewer
3
2. ChipViewer Strategy of Chip Array Design
(Test version)
4
3. ChipViewer De-Convolution of The Virtual
Tiling Chip
In the tiling chip, probes are randomly
allocated.
Raw
De-convoluted
5
4. ChipViewer Zoom in the virtual chip
6
5. ChipViewer Histograms of Hybridization
Signals in Genomic Orders
Find to find an annotated gene Mining to detect
expressing regions Specially novel
regions Fetch to fetch a regional sequence
7
6. ChipViewer Histogram hybridization and gene
annotation
8
7. ChipViewer Normalization Linear regression
normalization
y a x b
After normalization a 1 and b 0 for two sets
of chip data
y a x b
a
b
y log (intensity of the chip to be
normalized) x log (intensity of the standard
chip) a slope b intercept
9
8. ChipViewer Normalization Linear regression
normalization 2
Two sets of data different to each other
10
9. ChipViewer Normalization Distribution
normalization
Use a standard distribution or other chips
distribution model to normalize a set of data
Normalization based on the distribution of the
2nd chip
11
10. ChipViewer Normalization Comparing the 2
models
Linear regression model
Distribution based model
12
11. ChipViewer Statistic Methods and
Algorithms Weight Statistical
analyses

• Cramer von Mises test ( two tails)
• Kolmogorov-Smirnov test ( two tails)
• Weight (one tail)

n
Wt S (SN((i)) F((i))) d((i)) Rw
Wt / Wo
1
13
12. ChipViewer Weight Statistical analyses 2
1
NULL distribution
Chip distribution
Di
DKS
Distribution of Selected probes
Wt
0
x
14
13. ChipViewer Relative Weight Statistical
analyses 3
1
NULL distribution
Chip distribution
Di
Dks (i)
Wt
Wo
DKS (NULL)
Distribution of Selected probes
Rw Wt / Wo
0
x
15
14. ChipViewer Weight Statistical analyses 4
chip
DKS
Samples
NULL distribution ? chip distribution
v p(v)
16
15. ChipViewer Weight Statistical analyses 4-2
Samples
chip
NULL distribution ? chip distribution
17
16. ChipViewer Detection of novel genes
Use Mining button to detect the expression
regions.
This gene is now annotated as At1g09645 in the
latest AGI genome.
Use Mining button to detect the expression
regions.
18
17. ChipViewer Detection of novel genes 2
From the latest AGI genome to the future
annotation
ATG
ATG
TAA
RAFL
Nothing in the latest AGI genome, but a new gene
At1g11765 will be there in the future version.
19
18. ChipViewer Mapping of transcriptional units
of ORFeome
From 2000v At1g09750 (MIPS) to the latest AGI
At1g09750
2000 v Annotation (MIPS)
The latest AGI Annotation
20
19. ChipViewer Mapping of transcriptional units
of ORFeome
21
20. ChipViewer Mapping of transcriptional units
of ORFeome
Schematic of the Identification of New Genes
22
21. ChipViewer Identification of sm RNAs
5GACCCAATAAGCCGCTCCGATTGGA 3
CCB5
23
22. ChipViewer Evidence for Genes with Expression
24
23. ChipViewer Sense and anti-sense
Discovery of Sense and anti-sense signals in
Plasmodium and Arabidopsis.
Sense
Anti-Sense
Signals of the gene PFC0110w in Plasmodium.
25
24. ChipViewer Data fetching
Detect
Fetch
Six frames
Sequence
26
25. ChipViewer Comparing multiple sets of data
There are 3 sets of the Plasmodium tiling chip
data respectively in red, blue and green bars.
---- The Plasmodium tiling chip is a perfect
match only chip
27
26. ChipViewer Comparing multiple sets of data
(cluster tracks)
There are 9 sets of the chip data .
28
27. ChipViewer Outputs

• Output of de-convoluted data ( raw/normalized)
• Output of expression proportion output of the
  signal proportion of exon, intron and intragenic
  regions.
• Output of expression data mining output of the
  expressing regions and their weights.
• Output of all or selected probe sets and genes
  expressing data (statistical weight).

29
Acknowlegements
Thanks to Dr. Ecker, Dr. Lim and others at
the Salk Institute Genomic Analysis Laboratory
for their suggestions, comments and supports, as
well as thanks to Dr. Winzeler and Dr.
Grunenfielder from Scripps Institute.
Huaming Chen