Living Hardware to Solve the Hamiltonian Path Problem

1
Living Hardware to Solve the Hamiltonian Path
Problem
Professors Dr. Malcolm Campbell and Dr. Laurie
Heyer
Students Oyinade Adefuye, Will DeLoache, Jim
Dickson, Andrew Martens, Amber Shoecraft, and
Mike Waters
2
The Hamiltonian Path Problem
3
Computational Complexity

• Millennium Problem
• PNP?
• Brute Force required

Does a Hamiltonian Path exist in this graph?
4
Why Should We Use Bacteria?
VS.
Adleman LM (1994). Science 266 (11) 1021-1024.
5
Flipping DNA with Hin/hixC
6
Using Hin/hixC to Solve the HPP
7
Using Hin/hixC to Solve the HPP
hixC Sites
8
Using Hin/hixC to Solve the HPP
9
Using Hin/hixC to Solve the HPP
10
Using Hin/hixC to Solve the HPP
Solved Hamiltonian Path
11
What Genes Can Be Split?
GFP before hixC insertion
12
What Genes Can Be Split?
GFP displaying hixC insertion point
13
Gene Splitter Software
Input
Output

• Generates 4 Primers (optimized for
  melting temperature).
• 2. Biobrick ends are added to primers.
• 3. Frameshift is eliminated.

1. Gene Sequence 2. Where do you want
your hixC site? 3. Pick an extra base to avoid
a frameshift
14
Gene-Splitter Output
Note Oligos are optimized for melting
temperatures.
15
Use GFP to Split RFP
Green Fluorescent Protein
Red Fluorescent Protein
16
Can We Detect A Solution?
17
True Positives
Elements in the shaded region can be arranged in
any order.
(Edges-Nodes1)
Number of True Positives (Edges-Nodes1)! 2
18
False Positives
Extra Edge
19
False Positives
PCR Fragment Length
PCR Fragment Length
20
Detection of True Positives
Total of Positives
of Nodes / of Edges
of True Positives Total of Positives
of Nodes / of Edges
21
How Many Plasmids Do We Need?
k actual number of occurrences ? expected
number of occurrences
? m plasmids solved permutations of edges
permutations of edges
Cumulative Poisson Distribution
P( of solutions k) 1 -
22
Starting Arrangement
4 Nodes 3 Edges
Probability of HPP Solution
Number of Flips
23
Where Are We Now?
24
First Bacterial Computer
Starting Arrangement
25
First Bacterial Computer
Starting Arrangement
Solved Arrangement
26
Future Directions
Split additional genes
Make more complex graphs
Solve other problems such as the Traveling
Salesperson Problem
27
Living Hardware to Solve the Hamiltonian Path
Problem
Collaborators at MWSU
Dr. Todd Eckdahl, Dr. Jeff Poet, Jordan
Baumgardner,Tom Crowley, Lane H. Heard, Nickolaus
Morton, Michelle Ritter, Jessica Treece, Matthew
Unzicker, Amanda Valencia
Additional Thanks Karen Acker, Davidson
College 07
Support Davidson College
The Duke Endowment HHMI
NSF Genome
Consortium for Active Teaching
James G. Martin Genomics Program
28
Extra Slides
29
Traveling Salesperson Problem
30
Processivity
Problem

• The nature of our construct requires a stable
  transcription mechanism that can read through
  multiple genes in vivo
• Initiation Complex vs. Elongation Complex
• Formal manipulation of gene expression (through
  promoter sequence and availability of accessory
  proteins) is out of the picture

Solution T7 bacteriophage RNA polymerase

• Highly processive single subunit viral
  polymerase which maintains processivity in vivo
  and in vitro

31
Path at 3 nodes / 3 edges HP- 1 12 23
32
Path at 4 nodes / 6 edges HP-1 12 24 43
33
Path 5 nodes / 8 edges HP -1 12 25 54 43
34
Path 6 nodes / 10 edgesHP-1 12 25 56 64 43
35
Path 7 nodes / 12 edgesHP-1 12 25 56 67 74 43
36
More Gene-Splitter Output
37
Promoter Tester

• RBSKanRBSTetRBSRFP
• Tested promoter-promoter tester-pSBIA7 on varying
  concentration plates

• Used Promoter Tester-pSB1A7 and Promoter
  Tester-pSB1A2 without promoters as control
• Further evidence that pSB1A7 isnt completely
  insulated

38
Promoters Tested

• Selected strong promoters that were also
  repressible from biobrick registry
• ompC porin (BBa_R0082)
• Lambda phage(BBa_R0051)
• pLac (BBa_R0010)
• Hybrid pLac (BBa_R0011)
• None of the promoters glowed red
• Rus (BBa_J3902) and CMV (BBa_J52034) not the
  parts that are listed in the registry

39
Splitting Kanamycin Nucleotidyltransferase

• Determined hixC site insertion at AA 125 in each
  monomer subunit
• -AA 190 is involved in catalysis
• -AA 195 and 208 are involved in Mg2 binding
• -Mutant Enzymes 190, 205, 210 all showed changes
  in mg2 binding from the WT
• -Substitution of AA 210 (conserved) reduced
  enzyme activity
• -AA 166 serves to catalyze reactions involving
  ATP
• -AA 44 is involved in ATP binding
• -AA 60 is involved in orientation of AA 44 and
  ATP binding
• -We did not consider any Amino Acids near the N
  or C terminus
• -Substitution of AA 190 caused 650-fold decrease
  in enzyme activity
• -We did not consider any residues near ß-sheets
  or ?-helices close to the active site because
  hydrogen bonding plays an active role in
  substrate stabilization and the polarity of our
  hix site could disrupt the secondary structure
  and therefore the hydrogen bonding ability of
  KNTase)

40
Plasmid Insulation

• Insulated plasmid was designed to block
  read-through transcription
• Read-through transcription without a promoter
• Tested a promoter-tester construct
• RBSKanRBSTetRBSRFP
• Plated on different concentrations of Kan, Tet,
  and Kan-Tet plates
• Growth in pSB1A7 was stunted
• No plate exhibited cell growth in uninsulated
  plasmid and cell death in the insulated plasmid

41
Tetracycline Resistance Protein

• Did not split
• Transmembrane protein
• Structure hasnt been crystallized
• determined by computer modeling
• Vital residues for resistance are in
  transmemebrane domains (efflux function)
• HixC inserted a periplasmic domains AA 37/38 and
  AA 299/300
• Cytoplasmic domains allow for interaction with N
  and C terminus

42
Splitting Cre Recombinase
43
What Genes Can Be Split?
GFP before hixC insertion
GFP displaying hixC insertion point
44
Gene Splitter Software