Data

BT 3041: Analysis and interpretation

of Biological Data
 Outline
• Types of data
• Data visualisation
 Types of Data
• Data Set is a collection of Data Objects
• Data Object: also called
– Record, point, vector, pattern, event, case,
sample, observation, entity
• Data object is represented by a set of
Attributes
• Attribute: is a property or a characteristic of
an object
 Types of attributes
• Qualitative
– Nominal: just a name.
• eg. pincodes, IDs, eye color
– Ordinal: information to order objects.
• Eg. {good, better, best}
• Quantitative
– Numerical value exists
• Eg. temperature, pH
 Quantitative attributes
• Discrete
– Binary
• Continuous
 Characteristics of Data sets
• Dimensionality
– Curse of dimensionality
– Dimensionality reduction
• Sparsity
– Only a small fraction of attributes are non-zero
• Resolution
– Converting continuous quantities to discrete ones
 Types of data sets
• Record Data
• Graph/network Data
• Ordered data
• Spatial, image and multimedia data
 Record data:
Same number of attributes
Data matrix

Document data Market basket data

TID Items
timeout

season
coach

game
score
team

1 Bread, Coke, Milk

ball

lost
pla

wi
n
y

2 Beer, Bread
Document 1 3 0 5 0 2 6 0 2 0 2 3 Beer, Coke, Diaper, Milk
Document 2 0 7 0 2 1 0 0 3 0 0 4 Beer, Bread, Diaper, Milk
Document 3 0 1 0 0 1 2 2 0 3 0 5 Coke, Diaper, Milk
 Graph/Network data
• Examples:
– Internet

– Signaling networks

– Molecular structures
 Ordered Data
• Video data

• Time series data

• Molecular sequence data

 Spatial/image data
• Examples:
– Maps
– Images

http://imgarcade.com/1/india-temperature-map/
Basic Statistical Descriptions of Data
• Motivation
– To better understand the data: central tendency and spread
• Data dispersion characteristics
– median, max, min, quantiles, outliers, variance, etc.

12 (Hans, Kamber, Pei 2013)

 Measuring the Central Tendency
1 n
 x
• Mean (algebraic measure) (sample vs. population):
x
n
 x
i 1
i 
N
Note: n is sample size and N is population size. n

– Weighted arithmetic mean: w x i i

x i 1
n
– Trimmed mean: chopping extreme values
w
i 1
i

• Median:
– Middle value if odd number of values, or average
of the middle two values otherwise
Median
interval
• Mode
– Value that occurs most frequently in the data
– Unimodal, bimodal, trimodal
 Symmetric vs. Skewed
Data
• Median, mean and mode of symmetric, symmetric
positively and negatively skewed data

positively skewed negatively skewed

14
February 3, 2023 Data Mining: Concepts and Techniques
 Measuring the Dispersion of Data
• Quartiles, outliers and boxplots
– Quartiles:
• Q1 (25th percentile): The first quartile (Q1) is defined as the
middle number between the smallest number and the
median of the data set.
• The second quartile (Q2) is the median of the data.
• Q3 (75th percentile): The third quartile (Q3) is the middle
value between the median and the highest value of the
data set.
– Inter-quartile range: IQR = Q3 – Q1
– Five number summary: min, Q1, median, Q3, max
 Boxplot
– Boxplot: ends of the box are the quartiles; median
is marked; add whiskers, and plot outliers
individually
– Outlier: usually, a value higher/lower than 1.5 x
IQR
 Boxplot Analysis
• Five-number summary of a distribution
– Minimum, Q1, Median, Q3, Maximum
• Boxplot
– Data is represented with a box
– The ends of the box are at the first and third
quartiles, i.e., the height of the box is
Inter Quartile Range (IQR)
– The median is marked by a line within the box
– Whiskers: two lines outside the box extended
to Minimum and Maximum
– Outliers: points beyond a specified outlier
threshold, plotted individually
17
 Properties of Normal Distribution Curve

• The normal (distribution) curve

– From μ–σ to μ+σ: contains about 68% of the measurements
(μ: mean, σ: standard deviation)
– From μ–2σ to μ+2σ: contains about 95% of it
– From μ–3σ to μ+3σ: contains about 99.7% of it

18
Boxplot for normal distribution
 Visualization of Data Dispersion: 3-D Boxplots

20
February 3, 2023 Data Mining: Concepts and Techniques
 Variance
• Variance and standard deviation (sample: s, population: σ)
– Variance: (algebraic, scalable computation)
n n
1 1
      x  2
2 2 2
( xi ) i
N i 1 N i 1

– Standard deviation σ is the square root of variance or σ2

 Graphic Displays of Basic Statistical Descriptions

• Boxplot: graphic display of five-number summary

• Histogram: x-axis are values, y-axis represents frequencies
• Quantile plot: each value xi is paired with fi indicating that
approximately 100 fi % of data are  xi
• Scatter plot: each pair of values is a pair of coordinates and
plotted as points in the plane

22
 Histogram Analysis
• Histogram: Graph display of tabulated
40
frequencies, shown as bars
• It shows what proportion of cases fall35
into each of several categories 30

• Differs from a bar chart in that it is the25

area of the bar that denotes the 20
value, not the height as in bar charts,15
a crucial distinction when the 10
categories are not of uniform width
5
• The categories are usually specified as
0
non-overlapping intervals of some 10000 30000 50000 70000 90000

variable. The categories (bars) must

be adjacent
23
 Histograms Often Tell More than Boxplots

 The two histograms

shown in the left may
have the same boxplot
representation
 The same values for:
min, Q1, median, Q3,
max
 But they have rather
different data
distributions

24
 Quantile Plot
• Displays all of the data (allowing the user to assess both the
overall behavior and unusual occurrences)
• Plots quantile information
– For a data xi data sorted in increasing order, fi indicates that
approximately 100 fi% of the data are below or equal to the
value xi

25
Data Mining: Concepts and Techniques
 Scatter plot
• Provides a first look at bivariate data to see clusters of points,
outliers, etc
• Each pair of values is treated as a pair of coordinates and
plotted as points in the plane

26
Data Visualization
 Data Visualization
• Why data visualization?
– Gain insight into an information space by mapping
data onto graphical primitives
– Provide qualitative overview of large data sets
– Search for patterns, trends, structure, irregularities,
relationships among data
– Help find interesting regions and suitable parameters
for further quantitative analysis

28
VISUALIZING IN LOW DIMENSIONS
Pixel-Oriented Visualization Techniques
• 1D
Solar spectrum

• 2D

Microarray data

30
 3D visualization

Interactive visualization of three-dimensional

biological models in a CAVE. http://www.jvrb.org/past-issues/6.2009/2257
 Histogram
• Ex: Iris data
• 3 kinds of Iris flowers
• 4 attributes: petal length/width, sepal
length/width
• 50 samples per each flower type
Iris data: histograms of individual
attributes
 Pareto Chart
• A Pareto Chart is a graph that indicates:
– the frequency of causes of a certain
phenomenon,
– the cumulative impact of the causes.

Pareto Charts are useful to find the causes to

prioritize in order to observe the greatest overall
improvement.
They consist of a line graph and a bar graph
Pareto histogram

Frequencies
are sorted
Scatter plots
Correlation from scatter plots

Pearson Correlation Coefficient

Scatter plot in 3D
VISUALIZING SPATIO-TEMPORAL
DATA
Contour plot
Surface Plot
 Vector field plots

2D

A vector is depicted at every point

 Vector field plots

3D

A vector is depicted at every point

VISUALIZING IN HIGH DIMENSIONS
Parallel Coordinates plot
Star Coordinates

Sw – sepal width
sw Sl – sepal length
Pl – petal length
Pw – petal width
pl sl
pw
Chernoff faces

SL = size of face
SW = forehead/jaw ratio
PL = shape of forehead
PW = shape of jaw
A more intricate use of Chernoff faces
 Multiscale visualization

(Shimabukuro et al 2004)
Circle segment display
 CIRCOS

http://jura.wi.mit.edu/bio/education/hot_topi
http://circos.ca/intro/published_images/ cs/Circos/Circos.pdf
 Cone trees
• Display hierarchical data as cones
• Root node = apex; children = around the base
• Nodes are transparent so that you see the
nodes in the background
• Cones lower in the hierarchy are progressively
smaller
• If you click on a node, the node and the entire
path from the root node are highlighted
A Cone Tree
Another cone tree!