Utilizing Fuzzy Logic for Gene Sequence Construction from Sub Sequences and Characteristic Genome Derivation and Assembly

1
Utilizing Fuzzy Logic for Gene Sequence
Construction from Sub Sequences and
Characteristic Genome Derivation and Assembly
2
Contents

• Team
• Bioinformatics
• Genome Sequencing
• Research Problem
• Fuzzy Logic
• Ongoing Work
• Future Work

3
Team

• Advisors
• PI Dr. Gregory Vert (Dept. of Computer Science,
  University of Nevada Reno)
• Co-PI Dr. Alison Murray (Desert Research
  Institute, Reno)
• Co-PI Dr. Monica Nicolescu (Dept. of Computer
  Science, University of Nevada Reno)
• Student
• Sara Nasser (Dept. of Computer Science,
  University of Nevada Reno)

4
Bioinformatics- Genome Sequencing

• Genome sequencing is figuring out the order of
  DNA nucleotides, or bases, in a genomethe order
  of As, Cs, Gs, and Ts that make up an organism's
  DNA.
• Sequencing the genome is an important step
  towards understanding it.
• The whole genome can't be sequenced all at once
  because available methods of DNA sequencing can
  only handle short stretches of DNA at a time.

5
Genome Sequencing

• Much of the work involved in sequencing lies in
  putting together this giant biological jigsaw
  puzzle.
• Various problems occur such as
• Errors in reading
• Flips

6
Shot-Gun Sequencing

• The "whole-genome shotgun" method, involves
  breaking the genome up into small pieces,
  sequencing the pieces, and reassembling the
  pieces into the full genome sequence.

7
Environmental Genomics

• Multiple sequence alignment is an important first
  step in many bioinformatics applications such as
  structure prediction, phylogenetic analysis and
  detection of key functional residues.
• The accuracy of these methods relies heavily on
  the quality of the underlying alignment.

1
8
Multiple Sequence Alignment

• The traditional multiple sequence alignment
  problem is NP-hard, which means that it is
  impossible to solve for more than a few sequences
  1.
• In order to align a large number of sequences,
  many different approaches have been developed.

9
Tools and Techniques

• MUMMER
• Phrap, Phred, Consed
• TIGR
• The Smith-Waterman Algorithm
• Tree-Based Algorithms

10
Meta-genomics

• Meta-genomics is the application of modern
  genomics techniques to the study of communities
  of microbial organisms directly in their natural
  environments, bypassing the need for isolation
  and lab cultivation of individual species.

2
11
Meta-genomics

• Bacteria can often have minor variations in their
  DNA that can result in different metabolic
  characteristics.
• The differences can make it difficult to classify
  bacteria taxonomically.
• What has been needed is a method of creating a
  characteristic representation (characteristic
  genome) from the sub sequences of DNA found in
  several sub variant of a bacteria of the same
  species.
• Such genome could be used for more efficient
  classification at a molecular level through the
  process of controlled generalization.

12
Research Goals

• Given a collection of nucleotide sequences from
  multiple organisms, develop techniques based on
  fuzzy set theory and other methods for assembly
  of the sequences into the original full genome
  for each organism.
• Using the above techniques to develop a
  generalized approach for creating a
  characteristic genome that represents a
  generalization of the original organisms that
  donated sequence data.

13
The Data

• SYM (Original Raw Data)
• Contains 302K Sequences
• Average length of 450 base pairs (bp)
• It was obtained from a community of bacteria
• There is an estimated of 100 organisms
• Lets say, for example 75 of data is repeated, we
  still need to reassemble a sequence of 33
  Million bp

14
Motivation

• Current tools could not solve the problem
• Complexity of the dataset, since they are from
  same species.
• Sequencing environmental genomes, not a single
  organism.
• Limited tools that sequence environmental
  genomes.
• Algorithm
• Underlying algorithm determines the accuracy of
  match.
• Performance can be highly improved.
• Interfaces could be improved.

15
Problem

• Genome assembly is a O(2k) problem.
• Using Dynamic Programming it can be reduced.
• Example in seconds
• Assembly that takes around 1125899906842624
  seconds to can be reduced to 2500 seconds!

16
A Start

• We divide the problem in two steps
• Acquiring subsets such that each subset
  represents an organism
• Assembling this into a characteristic genome
  sequence
• The above two steps to be obtained by
• Clustering
• Assembly

17
Steps

Clustering
Assembly
Raw Data
Assembled Sequences
Characteristic Genome
18
The Data

• CAVEEG (Cleaned Dataset)
• Contains 128K Sequences
• Assembled Singletons
• Length ranges from 200bp-1000bp

19
D2 Cluster

• It is a software for clustering genome sequences
• The technique is based on distance.

20
Clustering with D2 Cluster

• Clustering was performed on 128K CAVEEG Dataset
• One dataset with 100K Sequences was obtained
• Majority of the data falls into one cluster
• This makes the process of separating organisms
  hard
• The clustering/assembly failed to assemble the
  sequences (the number of organisms were estimated
  manually and compared)

21
Problem with D2 clustering

• Does not look for contigs
• Ex A same cluster may have
• AATGCGTATTCGATGCGC
• CATACTTAGTCGATC AG
• When we assemble we desire
• AATGCGTATTCGATGCGC
• TGCGCATCGTATCG

22
Problems

• Since data is closely related the clustering
  technique assigns them to same cluster.
• Existing tools are unable to assemble the data
  correctly.
• The clustering software can only perform one
  round of clustering.

23
Ongoing Work

• Genome assembly using dynamic programming
• Uses Longest Common Sub-Sequence
• LCS is commonly used (ex Mummer)
• We added restrictions
• Enforce strict matches
• Encoding of data

24
Results
25
Then

• We added clustering.
• Instead of comparing each sequence with each
  other we can compare them with a group.
• Faster, less number of comparisons.

26
Clustering
3
27
Comparison of Clustering
28
Performance Comparison
29
How much does it matter?

• Obtaining an exact full length sequence it not
  essential
• A sequence that is very close to the original is
  desired

30
Snapshot 1
31
Snapshot 2
32
Fuzzy Logic

• Fuzzy Logic has been used extensively in
  approximate string matching using distance
  measures, etc.
• However, very little work has been done in
  application of building genomes from subsequences
  of nucleotides.
• The concept of similarity and application of
  fuzzy logic will be defined which is a relatively
  new area in nucleotide sequencing.

33
In Future..

• Compare technique with Phrap (alignment software,
  Mummer)
• Improve clustering
• Define Similarity using Fuzzy Logic
• Define Dissimilarity
• Parallelize the process

34
References

• 1 http//bioinformatics.oxfordjournals.org/cgi/c
  ontent/full/21/8/1408FIG1
• accessed May, 2006.
• 2 DeLong EF (2002) Microbial population
  genomics and ecology. Curr Opin Microbiol 5
  520524.
• 3 http//www.togaware.com/datamining/survivor/km
  eans04.png

35
Questions