A new revisited compression technique through innovative partition group binary

INTERNATIONALComputer Engineering and Technology ENGINEERING
International Journal of JOURNAL OF COMPUTER (IJCET), ISSN 0976-
6367(Print), ISSN 0976 – 6375(Online) Volume 4, Issue 2, March – April (2013), © IAEME
& TECHNOLOGY (IJCET)

ISSN 0976 – 6367(Print)
ISSN 0976 – 6375(Online) IJCET
Volume 4, Issue 2, March – April (2013), pp. 94-101
© IAEME: www.iaeme.com/ijcet.asp
Journal Impact Factor (2013): 6.1302 (Calculated by GISI)
©IAEME
www.jifactor.com

A NEW REVISITED COMPRESSION TECHNIQUE THROUGH
INNOVATIVE PARTITION GROUP BINARY COMPRESSION :A
NOVEL APPROACH

V. Hari Prasad
Sphoorthy Engineering college, A.P

ABSTRACT

Day by day the ratio of living organisms are increased and its accumulation in the
biological databases creating a major task. In this connection some of the classical algorithms
are strived into the world and fails to compress genetic sequences due to the encrypted
alphabets. Existing substitution techniques will work on repetitive and non repetitiveness of
bases of DNA and they achieve Best compression rates if any sequence may contain tandem
repeats. But the category of grasses like maize and rice contains very less repeats (nearer of
27) over 67789 total bps. By working existing techniques on such sequences the results are
not bountiful and running on worst case comparisons. This paper introduces a novel
Innovative partition group Binary compression technique yields first art compression rates
which is far better than existing techniques. This algorithm is developed based on
comparative study of existing algorithms and which is more applicable for non tandem
repeats of DNA sequences in genomes.

Keywords: Dnabit compress, Genbit compress, Huffbit compress, Encode, Decode

I. INTRODUCTION

The human genome was finally deciphered! In other words, scientists have succeeded
in reading the chain of more than 3 billion base pairs that constitute the DNA molecule of
humans; this process is called, sequencing. That daunting task required new analytical
methods created by bioinformatics .Bio informatics is nothing but information technology is
applied in terms of biological databases. To maintain such huge databases we require
efficient Bio informatics computational tools to store and process data in a more efficient
94

International Journal of Computer Engineering and Technology (IJCET), ISSN 0976-

way. Today more and more DNA sequences are storing in biological databases so the size of
the databases is growing in an exponential manner. Thus need of compression arises and it is
becoming a vital challenge, In this connection many universal algorithms are came into the
existence but they fails to compress genetic sequences due to the specificity of “text”. Some
classical algorithms are also introduced but they are running on negative compression rates.
Compression may be in two flavors one is lossless and other is loss. Loss compression can be
applicable for multimedia applications like image, audio and video. In multimedia
applications if we remove some unused pixels also resultant may not variant like removing
noise from audio or removing unnecessary pixels from images But Text compression is
always loss less even after decoding the entire encoded text we have to retain its original
property. DNA can be encoded in four letter alphabets like text {A, C, G, and T}.Thus each
Base of symbol (Base) can be represented by two bits. . General purpose compression
algorithms do not perform well with biological sequences. Giancarlo et al. [1][2] have
provided a review of compression algorithms designed for biological sequences. Finding the
characteristics and comparing Genomes is a major task (Koonin 1999[3]; Wooley 1999[4]).
In mathematical point of view, compression implies understanding and comprehension (Li
and Vitanyi 1998) [5]. Compression is a great tool for Genome comparison and for studying
various properties of Genomes. DNA sequences, which encode life should be compressible.
It is well known that DNA sequences in higher eukaryotes contain many tandem repeats, and
essentials genes (like rRNAs) have many copies. It is also proved that genes duplicate
themselves sometimes for evolutionary purposes. All these facts conclude that DNA
sequences should be compressible. The compression of DNA sequences is not an easy task.
(Grumback and Tahi 1994[6], Rivals et al. 1995 [7]; Chen et al. 2000 [8]) DNA sequences
consists of only four nucleotides bases {a,c,g,t}. Two bits are enough to store each base. The
standard compression software’s such as “compress”, “gzip”, “bzip2”, “winzip” expanded the
DNA genome file more than compressing it.
Most of the Existing software tools worked well for English text compression (Bell et
al. 1990[9]) but not for DNA Genomes. There are many text compression algorithms
available having quite a good compression ratio. But they have not been proved well for
compressing DNA sequences as the algorithm does not incorporate the characteristics of
DNA sequences even though DNA sequences can be represented in simple text form.DNA
sequences are comprised of just four different bases labeled A, T, C, and G (for adenine,
thymine, cytosine, and guanine respectively). T pairs with A, and G pairs with C. Each base
can be represented in computer code by a two character binary digit, two bits in other words,
A (00), C (01), G (10), and T (11). At first glance, one might imagine that this is the most
efficient way to store DNA sequences. Like the binary alphabet {0, 1} used in computers, the
four-letter alphabet of DNA {A, T, C, and G} can encode messages of arbitrary complexity
when encoded into long sequences.

A. Plan of the paper
This paper is organized as follows. Section 2 describes general compression algorithms.
Section 3 describes related existing algorithms to compress genome data. Section 4
describes proposed algorithms analysis how it is better one than existing techniques. Section
5 describes comparative study on a sample sequence. Section 6 is concluding with future
work.

95


II. GENERAL COMPRESSION ALGORITHMS
The compression of DNA sequences is considered as one of the most challenging
tasks in the field of data compression. In this connection the very first DNA compression and
its subsequent algorithms BioCompress[10] and BioCompress-2[11] detects exact repeats and
complementary palindromes located in the sequence and Encode the factor by the size
representation( l, p ) where l is the length of the factor and p is the position of its first
occurrence .If the size is greater the factor then use two bit encoding . More memory
references will require decoding the same, so the performance may degrade.
DnaPack[12] which uses hamming distance for the repeats and complementary palindromes
and it is implemented by dynamic programming approach. So that it is not simple in design
and it will require more time to execute and require more memory requirements also. The
algorithm achieves a compression rate in an average of 1.6602.
DNAcompress [13] will work on approximation of repeats if number of tandem repeats more
it saves bits to encode if not discard. Non repeated sequences will be appended to the
sequence at the end. This algorithm achieves a compression rate only 1.72 bits per base. If
there is no tandem repeat in the sequence it may run in worst case..DNASequitur is a
grammar based compression algorithm for DNA sequences which infers a context free
grammar to represent input data..Designing a CFG for the given input data may leads to
redundancies and constructing type2 language corresponding the grammar also leads to
ambiguity. The Lossless segment based compression enables part by part decompression by
introducing non base character so that it will save memory requirements but it is applicable
well on repeating sequences are more and more in the sequence. If such sequences like AT-
rich DNA, which constitutes a distinct fraction of the cellular DNA of the archaebacterium
Methanococcus voltae, consists of non-repetitive sequences, so part by part decompression is
little bit tedious.
III. RELATED EXISTING ALGORITHMS
Compression methods are fall into two categories.
• Statistical methods which compress data by replacing the shorter code. Huffman
code comes under this category and it’s not suitable larger sequences.
• Dictionary based replacing larger strings by shorter code. Cfact which searches the
longest exact matching repeat using tree data structures
Based on the above methods some loss less compression algorithms
strived based on two bits encoding schemes i.e. A(00),C(01),G(10) and T(11). HUFFBIT
COMPRESS [14], GENBITCOMPRESS [15] algorithms are explained performance analysis
(Best, Avg and Worst) based on repetitive and non repetitive bases of DNA and computed
results..Suppose in the given sequence more and more tandem repeats are there then [14],[15]
will run in Beast case and achieves bountiful compression ratios, if not the same may run in
worst case and achieves in an average 2.323 bits per bases[14],[15].Our proposed algorithm
PGBC(Partitioned group binary compression) techniques will achieves best compression
ration even the given DNA sequence may doesn’t contain tandem repeats or very less tandem
repeats which can consider as minute in a very larger sequence like category of maze and rice
grass sequences Our proposed algorithms are better suitable for non-repetitive DNA
sequences in genome and which is achieving in an average of 1.333 bits per bases which is
far better than all existing techniques.

96


IV. PROPOSED ALGORITHM
Our proposed algorithms are developed based on the comparative study of existing
techniques and we are starching on non-repetitiveness of DNA sequences. Existing
techniques are run in worst case if any DNA sequences contain no tandem repeats. Here we
are working with worst-case scenarios and achieves better compression rates in terms bits per
symbols.
A. Idea behind the algorithm
Every DNA sequence contain {A, C, G, T} nucleotides where each literal is named as
BASE and encoded in two bits as follows
A=00, C=01, G=10 and T=11.
Compression ratio is calculated encoded bits per Bases.
Compression Ratio = Encoded Bits/Bases.
B. Plan of work
Here we took and input sequence as sample DNA (which doesn’t contain any tandem
repeats) of length n and divides it into n/4 fragments (where each fragment contain four
bases i.e. A, C, G and T). In Encoding process every six fragments can grouped as partition
(P) which contains two sub partitions (Fh and Sh ) . We can substitute its equivalent binary
bits before making sub partitions and later we can group it into single main partition set. (Gs).
In decoding process we can do the reverse or encoding to retain loss less DNA property.
Finally we will calculate will group all the partition the equations are total number of
encoded bits by grouping all the part ions.
Suppose if we took sample sequence of DNA which contain 72 bases then by applying
PGBC techniques it will fragmented into n/4 i.e. 18 fragments , 3 partitions which will
contain 6 sub partitions and then grouped it into single main partition. This will represent
number of encoded bits in the given sequence. Our PGBC technique work as follows

Partition set (Ps) can calculate as follows.

Group partition set can calculate as follows

Here Gs will represent the binary equivalent numeric (nearest to integer) in terms of Bytes
storage. (Suppose if we will implement the technique in C language unsigned int will require
4 bytes of storage).Here m=n i.e. length of the given sequence.

97


Total Number of Encoded group bits are calculated as follows.

Finally compression Ratio can calculate as follows.

C. Analysis
Length of the given sequence N = 72 Bases.
Then possibly 3 part ion sets which will include 6 sub partitions (3-Fh and 3-Sh) and
finally we can group all partitions into a Group partition set.

Ps=P1 + P2 + P3.
we will calculate binary equivalent numeric(which is nearer to integer) value for each
partition in terms of Bytes storage(suppose if we will store each partition in C we will
require two bytes if we can accommodate on int if not we can go for unsigned long).
Gs=Ps (p1 + p2 + p3)

Total Number of Encoded Bits
Eg b= Gs= Ps
Every partition may contain 24 bases so it may not fit in integer so that we can store in
unsigned long. So totally our sequence is divided into three partitions and grouped as one
set .So totally we require 12 bytes to store.

=4 + 4 + 4 = 12 Bytes (96 bits)
Finally we calculate Compression Ratio as follows.

= 96 / 72
= 1.333 (bits per Bases.)
Encoding and decoding algorithms for DNA compression is as follows.
D. Encoding Algorithm
INP: input String
OPS: Encoded String
PROCEDURE ENCODE
Begin
• Group INS into equivalent fragments as four bases
• Generate all possible combinations of DNA and it will contain non- repetitive (our INS
assumed as no tandem repeats).

98


• Group six fragments into partition set which will consist of two subpartions.
• Assign binary bits(0&1) for every base of DNA like
A=00, C=01, G=10 and T=11
• Calculate Gs for every Ps in INP till eof INP

• Calculate Egb for every Gs till eof INP

• Repeat the steps 4 and 5 until the length of the INP
• Transfer the sequence Egb to the output string i.e. OPS String.
End.

E. Decoding Algorithm
INP: input String
OPS: Decoded String
PROCEDURE DECODE

Begin
• Generate all possible combinations of (A,C,G,T)
• Read the binary data of each sub partition from OPS and assign the two bits by
equivalent Base s (00=A,01=C,10=G and 11=T) and then store it in an array till eof
• Repeat step 2 until eof INS is reached and calculate Dgb and Ds in the reverse
process..
• Transfer the sequence Db to the input String i.e. INP
End.
V. EXAMPLE AND COMPARISON
Let us consider the sequence.
Sequence1:
ACGT GCGC GATC GCCT GCTA GGCG TACG TCGC AGGC GATC GATG TGCT
AGAT CAGA TGAC TCAG
TGCA CGAT.
Sequence length (no of bases) = 72.
Bytes required to store in a text file = 72 Bytes.
The above sequence doesn’t contain tandem repeats so existing algorithms like Huffbit
compress,Genbit Compress and Dnabit compress may run on worst case and require more
bits to encode the sequence.
Huffbit,GenBit and Dna compress =162 bits(2.25)
Genbit Compress (Tool based) = 160 bits (2.23)
PGBC Technique (Compression) = 96 bits (1.333)

99


VI. CONCLUSION AND FUTURE WORK

By using of our algorithm we can encode every base by 1.33 bits .By applying of ours
we are saving nearer of 8 bytes to encode the given sequence, compression may vary with
size of the sequence. So our technique is far better than existing ones and we can apply this
technique on non repetitive DNA sequences of genomes .If the given sequence can contain
tandem repeats also our technique will achieve same compression rate in an average. In
addition to that existing techniques uses dynamic programming to compress the sequence
which is complex in implementation and time consuming. Our technique is implemented
without dynamic programming approach, so it is simple and fast. The simplicity of this will
reduce the complexity in processing and definitely it will be the invaluable tool in Bio
informatics era. Our algorithm can be extended to any tool based approach.

ACKNOWLEDGEMENTS

We would like thank other members of the Bio –Informatics teams (Faculty of CSE
and IT) at Sphoorthy Engineering college Nadergul(V),R.R Dist,Hyderabad. I am very much
greatful to my mother and father V.subba lakshmamma and V.C Obanna in every path of my
success. Last but not least students of CSE in sphoorthy for sharing their ideas with us in
refining of our architecture

REFERENCES

[1] E Schrodinger. Cambridge University Press: Cambridge, UK, 1944.[PMID:
15985324]
[2] R Giancarlo et al. A synopsis Bioinformatics 25:1575 (2009) [PMID:19251772]
[3] EV Koonin. Bioinformatics 15: 265 (1999)
[4] JC Wooley. J.Comput.Biol 6: 459 (1999) [PMID: 10582579]
[5] CH Bennett et al. IEEE Trans.Inform.Theory 44: 4 (1998)
[6] S Grumbach & F Tahi. Journal of Information Processing and Management 30(6):
875 (1994)
[7] E Rivals et al. A guaranteed compression scheme for repetitive DNA sequences.
LIFL, Lille I University, technical report IT-285 (1995)
[8] X Chen et al. A compression algorithm for DNA sequences and its applications in
Genome comparison. In Proceedings of the Fourth Annual International Conference on
Computational Molecular Biology, Tokyo, Japan, April 8-11, 2000. [PMID: 11072342]
[9] TC Bell et al. Newyork:Prentice Hall (1990)
[10] J Ziv & A Lempel. IEEE Trans. Inf. Theory 23: 337 (1977)
[11] A Grumbach & F Tahi. In Proceedings of the IEEE Data [12]
[12] DNA compression is challenge is revisited Beshad Behajadi
[13] Allam AppaRao.In proceedings of the Bio medical Informatics Journal
[2011].DNABIT compress-compression of DNA sequences
[14] Allam AppaRao.In proceedings of the JATIT journal computationalf Biology and
Bio Informatics:[2009].HuffBit compress-compression of DNA using extended binary trees
[15] Allam AppaRao.In proceedings of the JATIT journal computational Biology and Bio
Informatics:[2011].Genbit compress-compression of DNA sequences.

100


AUTHORS’ INFORMATION

V Hari Prasad, Assoc. professor, B.Tech CSE from JNTU University,
Anantapur, M.Tech CSE from JNTUCEH,HYD and pursuing research in the
area of Bio Informatics at JNTU KAKINADA, A.P as a External Research
scholar in CSE .He has 10 years of teaching experience in various
Engineerig colleges. Presently He is heading the CSE Dept at Sphoorthy
Engineering college ,Nadergul(V),Hyd. He is a Life Member of MISTE and
Member of IEEE and UGC-NET qualified.He presented papers at International & National
conferences on various domains. His interested areas are Bio Informatics, Databases, and
Artificial Intelligence.

101

A new revisited compression technique through innovative partition group binary

More Related Content

What's hot (19)

Viewers also liked (8)

Similar to A new revisited compression technique through innovative partition group binary (20)

More from IAEME Publication (20)

A new revisited compression technique through innovative partition group binary