Scribe: S. Oak

Today's Topics

• Introduction to Fragment Assembly.
• Sequencing --- To obtain a string of bases that make up a DNA molecule.
• Target (S) --- Long sequence resuliting from Fragment Assembly.
• Fragment --- Sequencing the whole DNA molecule is not possible due to its length. However a piece of the molecule can be used and sequenced in the canonical (5' ---> 3') direction for a certain length. Each such sequence is called a fragment.
• Fragment Assembly --- The whole DNA molecule is then deduced from these collection of fragments and hence the process is termed as fragment assembly.

 
• Guidance to Fragment Assembly.
• Overlaps --- Sometimes the end part of a fragment is similar to the beginning of the other.
• Approximate Size ---The approximate length of the target DNA molecule is known.
• Contigs --- connected pieces of the target DNA that we assemble.

• Result of Fragment Assembly.
• Consensus sequence (consensus) --- The consessus is obtained by taking the majority vote among all the bases in each column during assembly.
 
• Complications for Fragment Assembly.
• Base call errors --- The Base call errors comprise base substitutions, insertions and deletions in the fragments.
• Chimeras --- Chimeras arise when 2 regular fragments from distinct parts of the target molecule join end-to-end to form a fragment that is not a contiguous part of the target.
• Contamination --- Some fragments that do not have anything to do with the target molecule are the cause for contamination.This is due to contamination from the host or vector DNA.
• Unknown orientation --- Fragments can come from any DNA strands and hence the fact that a fragment is a substring of 1 strand is equivalent to the fact that its reverse complement is substring of other.
• Repeated regions --- All the overlaps involving repeats may give an accepatable answer and hence this complicates the assembly process.
• Alignment --- Imperfect alignment can result due to skip of  a base.
• Substrings --- One fragment can be a part of another or it may be contained in another.
• Lack of coverage --- Coverage at position i of the target is the number of fragments that cover that position. Coverage is difficult to compute because the actual positions of the fragments in the target is not known.
 
• The message arrival model was discussed to explain:
• Stochastic Process --- A counting process for a set of events occuring in time interval(0, ¥)
• Poisson Process --- A counting process for a set of events with constant rate.
 
• The Poisson distribution formulae were applied to give the expression for esimated number of apparent contigs as : p = ne^-n(l-t)/T

• The fraction of target sequence covered by exactly 'k' fragments is given by: r = n(e^-cc^k)/k!

Today's Sources

• Setubal and Meidanis: Chapter 1 (Pages 15 - 21) and Chapter 4 (Pages 105 - 113).
• Salzberg, Searls, and Kasif: not used today.
• Introduction to Computational Biology : Maps, Sequences and genomes. By : Micheal Waterman. ( Pages 94 -98)
• Getting all fragments using the message arrival model.