Newcastle University/22 May 2008

Back to Calendar

Minutes

Minute Taker = Nina Nielson

Action Plan for next week

Group

•	Come up with a team logo and perhaps a flyer/poster if feeling artistic.

•	Find out if Streptococcus pneumoniae’s target for VncR/S is still “unknown” as it was in 2002 paper/book. This would make the species impossible to use!

•	 Find out for sure whether B. subtilis has the PapR/PlcR system like B. cereus.

•	Email Jan-W J. Can we distinguish between red and green being on, red being on and green being on? What are the individual peak wavelengths of the different available colours and are they distinguishable from one another?

•	Set up a calendar on the Wiki for absences in the summer to be entered.

•	Make sure all of the weeks work and the papers used are up on the Wiki.

•	Research the peptides and their parts.

Megan

•	Google “evolving logic gates for neural networks”.

•	Write characteristics for first layer.

•	Write very basic java doc for database structure (for a record collection)

•	Write java pseudo code for parts database.

•	Investigate more parts for the chosen peptides Nina

•	Google “evolving logic gates for neural networks”.

•	Write very basic java doc for database structure (for record collection)

•	Write pseudo code for constraints database.

•	Add a “types of parts” compatibility matrix to the “parts” one.

•	Meet with Matt (10:00 Tuesday) to go over RDF

Mark •	Register to iGem 2008

•	Put Blast result of different peptides of different strains of S. aureus on wiki.

•	Find the maximum connectivity of a node in a neural network.

•	Investigate S. aureus and L. monocytogene further. Identify more layer features.

The meeting

Time: 14:30 – 17:00

Attended by:

Neil Wipat – Supervisor

Jen Hallinan – Supervisor Morgan Taschuk – Project 5

Mark Wappett – Project 3 Megan Aylward – Project 1

Nina Nielsen – Project 2

Meeting Content

Agenda 1: Finalising the Biological concept

Megan Starts off the meeting stating that she, Mark and Nina have come up with feasible bacterial pathogens and respective individual peptides to detect. However, Clostridium difficile and Bacillus anthracis turned out not to be appropriate.

Nina Explains that Clostridium difficile does not have any direct evidence of a QSS. The LuxS system secretes the autoinducer AI-2, which is common to many bacterial species and cannot be a diagnostic target.

Mark Explains that Bacillus anthracis’ genome has only very recently been sequenced and not fully annotated yet. This has meant that retrieving information has been difficult.

Therefore the students have decided to not use Bacillus anthracis or Clostridium difficile.

Nina and Mark Have identified two other potential bacterial pathogens to detect: Listeria monocytogene and Bacillus cereus.

This brings the total number of pathogenic species detected to four:

1.	Staphylococcus aureus

2.	Streptococcus pneumoniae

3.	Listeria monocytogene

4.	Bacillus cereus

Neil suggests that as a team we could propose a second slant of the project being a pathogenic food sensor. (Since S. aureus, L. monocytogene and B. cereus can all be food related bacterial infections) However the diagonostic would then have to detect the concentration of bacteria not just the presence of it.

Megan draws list of parts on board (Fig. 1)

Fig.1 Table of first layer of parts

'''Bacterial species	S. pneumoniae	S. aureus	L. monocytogene	B. cereus'''

Peptide                  Pep27	    AIP	               L.AIP	     OS5

Receptor / TF	    VncS/R	   AgrC/A	       AgrC/A	     PlcR

Target?	                           AgrB	       L.AgrB	PlcR regulon

Megan Mentions that Pep27 has a 40% sequence similarity score when blasted and Neil approves of this figure. Mark Clarifies that both L. monocytogene and S. aureus use the AgrC/A two component system. However he has blasted the AIP’s and identified a considerable difference. Agenda 2: FPs

The fluorescent proteins will be the last layer of the table above (Fig. 1) as they are the output. However we need to define which colours represent what. Jen Points out that we need a default indicator to show that the system is working. I.e. green. However this leaves us with only mCherry and YFP left. Morgan suggests that we could have a low level GFP. This could be constitutive. For example, if one uses LuxI, a “glow” will be emitted rather than a fluorescent light. However this “glow” is very faint and would be difficult to observe.

Jen searches the registry for “Logic gates” or “Nand gates” parts. These provide more output combinations. However to work, we need to know if the colours we are working with can be distinguished when switched on in combination. Could we make an orange out of the yellow and red switched on together? What are the wavelengths of each of the FP’s? (Email JW-V).

Neil describes on the board what logic gates are. Draws below table on board (Fig. 2).

Fig. 2 Table of Logic Gate combinations

Combination	AND *	OR	XOR

0 + 0	0	0	0

0 + 1	0	1	1

1 + 0 	0	1	1

1 + 1 	1	1	0


 * The prefix N forming a “NAND” gate is the inverse. Therefore the results would be 1;1;1;0 for this column.

Agenda 3: Individual project progress

Project 1(and 2)

Megan Has been working with CellML this week to build models of the parts. Asks about what programming it would involve? Neil Explains that Java will be written on top of the CellML. Neil Draws a schematic of the design (Fig.3).

Fig. 3 Schematic of program design

Java Program

get_Model

get_Part

get_PartId

get_ModelId

•	It is important to make sure that the user doesn’t need to write SQL queries.

•	The interface doesn’t change at all.

•	Can change the database without affecting the interface.

•	Hand code your SQL query in Java (probably easiest way).

E.g. SELECT “name”

FROM “user”

•	Look at JDBC. JDBC passes the information as a string. It bundles the information as it iterates (like a for loop). Access supports this.

Morgan draws a schematic on the board of how the interface, server and Java classes relate (Fig. 4).

Fig. 4 How the components of the program relate.

Database1

Database2

•	Nina The above information applies to you too, although using RDF and/or CellML.

Megan and Nina

Exercise: Try to code up the above for a simple database such as a record collection. Consider song number, song name, song genre etc to make up the database.

All the students should now be doing a bit of coding everyday.

Project 3

Mark Has written up a Java code this week. He asks for practise suggestions. Jen suggests;

•	to pick a node and see what it turns on

•	do something with topology

•	Find the maximum connectivity of a node.

•	CellML will eventually be used as part of the fitness function. Used to fit parts together. But don’t have to worry about that now.

Megan Asks about constraints.

Are constraints such as part length entered into CellML? Are they values?

Neil Suggests that the group just focus on coding up the java.doc for now. Can be read only for now to be a little easier.

AOB

•	Neil is away next week

•	Jen suggests the “Mini-talks” should be held in two weeks time (05/06) when Neil is here. These should be 5 minute presentations, summarizing progress on our individual projects so far. Good practise for presenting skills.

•	We all get out badges and go to the pub…Woooooooo!!