A Standard Operating Procedure for
Systems Biology should adress at least the following items: |
|
|
|
|
|
|
|
|
1 |
Study object |
|
|
|
|
2 |
Study design |
|
|
|
|
3 |
Experimental conditions |
|
|
|
|
4 |
Parameters to be measured |
|
|
|
|
5 |
Measuring methodologies |
|
|
|
|
6 |
Assay and model
validation |
|
|
|
|
7 |
Modeling software |
|
|
|
|
8 |
Record keeping, data
collection/storage/access |
|
|
|
|
9 |
Reporting and publishing |
|
|
|
|
10 |
Project management |
|
|
|
|
|
|
|
|
|
|
|
|
|
The
following shows when and how often these were presented and discussed at the
standards workshop. |
|
Workshop
G/H on Standards in Systems Biology at ICSB 2004 in Heidelberg on 10/10/04 |
|
|
|
|
|
|
|
|
Speaker |
|
Subject |
|
Adressed in SOP item # |
|
|
Carsten
Kettner |
Experimental conditions |
|
|
|
|
|
|
3 |
|
|
|
|
on
Strenda |
Parameters under study |
|
4 |
|
|
|
|
|
Database characteristics |
|
7 and 8 |
|
|
|
|
|
Reporting methods |
|
9 |
|
|
|
Discussion
--------------- |
How can we agree to agree |
|
1 thru 10 |
|
|
|
|
|
|
Study same cells or
organism |
|
1 |
|
|
|
|
|
Use of model templates |
|
7, 8, 9 |
|
|
|
|
|
Unified methodology |
|
5, 6, 7 |
|
|
|
|
|
Intracellular conditions |
|
1, 3, 4 |
|
|
|
|
|
Reconsider past results |
|
|
|
|
|
|
Not adressed |
|
|
|
Dietmar Schomburg |
High-quality database and
thesaurus |
|
|
|
5, 6, 7, 8, 9 |
|
|
|
on
Brenda |
Chaotic data |
|
2, 3, 4, 5 |
|
|
|
|
|
Source and history of
cells |
|
1 |
|
|
|
Discussion
--------------- |
Atom-mapping info |
|
3, 4, 5 |
|
|
|
|
|
Thermodynamics info |
|
3, 4, 5 |
|
|
|
|
|
Enzyme units |
|
3, 4, 5 |
|
|
|
|
|
Protocols, reporting |
|
8, 9 |
|
|
|
|
|
Referee quality |
|
9 |
|
|
|
|
|
Range of experimental
conditions |
|
3 |
|
|
|
|
|
Value of in vivo comparisons |
|
|
|
|
|
|
2, 6 |
|
|
|
|
Uwe
Sauer |
Data authentication |
|
8 |
|
|
|
on
measuring flux |
Assay validation and
diversification |
|
6 |
|
|
|
|
|
No single method
sufficient |
|
Not adressed |
|
|
|
|
|
Highest quality not
always needed |
|
2, 5, 6 |
|
|
|
|
|
Data inconsistencies
become visible |
|
8, 9 |
|
|
|
|
|
Start with "low
quality" |
|
2, 5, 6 |
|
|
|
YSBN |
|
Strains and conditions |
|
1, 3 |
|
|
|
|
|
Raw data |
|
8 |
|
|
|
|
|
Simple physiology |
|
2, 3, 4 |
|
|
|
|
|
Data clustering |
|
6, 7 |
|
|
|
Discussion
--------------- |
Include genome info |
|
1, 4 |
|
|
|
|
|
Include flux data |
|
4 |
|
|
|
|
|
Try "grading"
data quality |
|
5, 6 |
|
|
|
|
|
Data variability and
statistics |
|
2, 5, 6, 8 |
|
|
|
|
|
Reporting and publishing
standards |
|
9 |
|
|
|
|
|
Protocol disambiguation |
|
|
|
|
|
|
2 |
|
|
|
|
Jildau
Bouwman |
Try aligning labs and
study objects |
|
|
|
1, 2, 3, 4, 5 |
|
|
|
on
standardization |
Allow for serendipity |
|
Not adressed |
|
|
|
|
|
Use "right"
sampling frequency |
|
3 |
|
|
|
|
|
Define experimental
conditions |
|
3, 5, 6, 7 |
|
|
|
Discussion |
Sampling frequency |
|
3 |
|
|
|
|
|
Selection of strains |
|
1 |
|
|
|
|
|
Keep strain number
minimal |
|
1 |
|
|
|
|
|
Use timetable when
switching strains |
|
1 |
|
|
|
|
|
Polymorphism in strain
sequence |
|
1 |
|
|
|
|
|
Incorporate other
disciplines' databases |
|
|
|
|
|
8 |
|
|
|
|
Workshop
G/H on Standards in Systems Biology at ICSB 2004 in Heidelberg on 10/10/04 |
|
|
|
|
|
|
|
|
Speaker |
|
Subject |
|
Adressed in SOP item # |
|
|
Jan-Marie
Francois |
Arrest all metabolic
activity |
|
|
|
|
|
|
3 |
|
|
|
|
on standardization |
Standardize extraction
methods |
|
3, 4 |
|
|
|
of metabolic |
Standardize assay methods |
|
5, 6 |
|
|
|
sampling |
|
Result dependent on
method |
|
5, 6, 7 |
|
|
|
|
|
Avoid cell lysis |
|
3, 4, 5, 6 |
|
|
|
|
|
|
Common method unlikely to
evolve |
|
2, 3, 4, 5, 6, 7 |
|
|
|
|
|
Differentiate between
free and bound metabolites |
|
4 |
|
|
|
Discussion
--------------- |
Add non-invasive methods |
|
5 |
|
|
|
|
|
Metabolite
compartmentalization |
|
3, 4, 5, 6 |
|
|
|
|
|
Use more than one method |
|
|
|
|
|
|
5 |
|
|
|
|
Ursula
Klingmueller |
Cell proliferation and
differentiation |
|
|
|
3, 4 |
|
|
|
on
SB standards |
Traditional methods
insufficient |
|
4, 5, 6 |
|
|
|
|
|
Mathematical modelling
approach |
|
6, 7 |
|
|
|
|
|
Resolution in time and
space needed |
|
4, 5 |
|
|
|
|
|
Biologists need
"reward" |
|
9, 10 |
|
|
|
|
|
Describe dynamic behavior |
|
4 |
|
|
|
|
|
Strict SOP's needed |
|
1 thru 10 |
|
|
|
|
|
Multidisciplinary
approach needed |
|
4, 5 |
|
|
|
|
|
Aim for highest data
quality |
|
2, 5, 6 |
|
|
|
|
|
Stimulate comm. between
modellers and experimenters |
|
10 |
|
|
|
Discussion
--------------- |
Make new start involving
all relevant disciplines |
|
10 |
|
|
|
|
|
Take the EU-wide
perspective |
|
1 thru 10 |
|
|
|
|
|
Compare
"biological" errors to "technological" errors |
|
5,6 |
|
|
|
|
|
Re-appraisal of
"old" data needed |
|
Not adressed |
|
|
|
|
|
What about
"industrial reality"? |
|
10 |
|
|
|
|
|
Even data of imperfect
quality can make models work |
|
5, 6, 7 |
|
|
|
|
|
Don't tell biologists
about their need for reward |
|
|
|
|
9, 10 |
|
|
|
|
Ursula
Kummer |
Facilitate IT-access for
non-nerds |
|
|
|
7, 8, 9, 10 |
|
|
|
on
Interfacing users |
Standardize computational
methods |
|
6, 7 |
|
|
|
with
modelling |
Sticking to GUI standards
often suffices for publication |
|
6, 7, 9 |
|
|
|
|
|
Use automation maximally,
mathematics minimally |
|
5, 6, 7 |
|
|
|
|
|
Selection of methods
should be rational |
|
3, 4, 5 |
|
|
|
|
|
Use web-based databases
to facilitate modelling |
|
8 |
|
|
|
Discussion
--------------- |
Streamline issues by
using SBML and GUI interface |
|
6, 7 |
|
|
|
|
|
Use models that can be
understood and interpreted |
|
6 |
|
|
|
|
|
Set required level of
mathematics as low as possible |
|
6, 7 |
|
|
|
|
|
Improve model
reproducibility |
|
6, 7 |
|
|
|
|
|
Use clearly identifiable
algorithms and software |
|
7 |
|
|
|
|
|
Create non-profit source
of available software |
|
7 |
|
|
|
|
|
Stick to agreed
nomenclature |
|
|
|
|
|
|
6, 7, 9 |
|
|
|
|
Andrew
Finney |
Use open-access, accepted
standard |
|
|
|
6, 7 |
|
|
|
on SBML |
|
SBML supported by over 60
software applications |
|
7 |
|
|
|
|
|
SBML stimulates
collaboration between groups |
|
10 |
|
|
|
|
|
Publication apportunities
increased by using SBML |
|
7, 9 |
|
|
|
|
|
SBML is extremely
flexible |
|
7 |
|
|
|
|
|
On-line validation
possible |
|
6 |
|
|
|
Discussion
--------------- |
Acronym usage
questionable, thesaurus needed |
|
7, 9 |
|
|
|
|
|
Introduce annotations in
SBML |
|
7 |
|
|
|
|
|
Terminology still cloudy |
|
7, 9 |
|
|
|
|
|
Use SBML for stochastic
models |
|
6, 7 |
|
|
|
|
|
Simulation not dependent
on SBML |
|
6, 7 |
|
|
|
|
|
Available software has
unspecified limitations |
|
6, 7 |
|
|
|
|
|
Divide test suites in
segments |
|
|
|
|
|
|
6, 7 |
|
|
|
|
Workshop
G/H on Standards in Systems Biology at ICSB 2004 in Heidelberg on 10/10/04 |
|
|
|
|
|
|
|
|
Speaker |
|
Subject |
|
|
|
|
|
|
|
|
Adressed in SOP item # |
|
|
Sef
Heijnen on |
Network structure
substantiation important |
|
|
|
|
|
4, 5, 6 |
|
|
|
|
modelling
kinetics |
Interactions are often
non-linear |
|
6 |
|
|
|
|
|
Limited controllability
of parameters |
|
4 |
|
|
|
|
|
Establish in-vivo parameters as well |
|
1, 2, 3, 4 |
|
|
|
|
|
Characterize steady-state
conditions |
|
3, 4 |
|
|
|
|
|
Perturbations generate
new data |
|
3, 4 |
|
|
|
|
|
Living systems never
reveal all potential data |
|
1 thru 5 |
|
|
|
|
|
Extrapolation from in-vitro to in-vivo is difficult |
|
2 thru 5 |
|
|
|
|
|
No "Grand
Theory" sought after |
|
2 |
|
|
|
|
|
Standards for kinetic
studies needed |
|
2 thru 5 |
|
|
|
|
|
Present parameters in a
normalized way |
|
4, 5, 6, 8, 9 |
|
|
|
|
|
Modelling achievable but
difficult |
|
6, 7 |
|
|
|
|
|
Cell-cycle averaging is
unavoidable |
|
2, 3, 4 |
|
|
|
|
|
Data-driven or
model-driven? |
|
2 |
|
|
|
Discussion
--------------- |
Selection of dynamic
state conditions |
|
2, 3, 4 |
|
|
|
|
|
Moving away from
approximation causes trouble |
|
5, 6 |
|
|
|
|
|
Approximations include
linearizations |
|
5, 6 |
|
|
|
|
|
New parameter-fitting
could help SB |
|
6, 7 |
|
|
|
|
|
Don't forget physiology |
|
2, 6 |
|
|
|
|
|
Do cultures really
exhibit steady-state conditions? |
|
|
|
|
2, 3, 4 |
|
|
|
|
Edda Klipp on |
Publications are crucial
for replicating experiments |
|
|
|
|
2 thru 5, 9 |
|
|
|
YSBN
modelling |
Difference between
prototype and reality |
|
6, 7 |
|
|
|
standards |
Compatibility issues |
|
1 thru 9 |
|
|
|
|
|
Parameter change outcome
should be realistic |
|
4 |
|
|
|
Discussion
--------------- |
Publications often
incomplete, template needed |
|
9 |
|
|
|
|
|
Make all data available
on the Web |
|
9 |
|
|
|
|
|
Specify/explain use of
symbols/acronyms |
|
9 |
|
|
|
|
|
Clarify use of
approximations |
|
6, 7, 8 |
|
|
|
|
|
|
Provide access to
software |
|
7 |
|
|
|
|
|
Justify choice of
similation parameters |
|
|
|
|
|
6, 7 |
|
|
|
|
Jacky
Snoep on |
Behavior is a function of
the components |
|
|
|
|
|
4, 6, 7 |
|
|
|
|
the
Silicon Cell |
Measure parameters
experimentally, in vivo too |
|
4, 5, 6 |
|
|
|
|
|
|
Unify model storage in
database |
|
8 |
|
|
|
|
|
Integrate various model
subsets |
|
6 |
|
|
|
|
|
Apply rigorous model
validation |
|
6 |
|
|
|
|
|
Detail kinetics of all
reaction steps |
|
4 |
|
|
|
Discussion
--------------- |
Whole-cell perturbation
important tool |
|
4 |
|
|
|
|
|
Handle signal
transduction data mathematically |
|
6 |
|
|
|
|
|
Include and link various
abstraction levels |
|
6 |
|
|
|
|
|
Agree on merging modules
into models |
|
|
|
|
|
6 |
|
|
|
|
|
|
|
|
|
|
|
|
|
The
following table shows the relative frequency of each of the items adressed
during the workshop |
|
|
|
SOP Items |
|
|
|
|
No of times mentioned |
|
Order of perceived
importance |
|
|
|
1 |
Study object |
|
16 |
|
|
item 6 |
most frequently mentioned |
|
|
2 |
Study design |
|
23 |
|
|
items 4 and 7 |
|
|
|
3 |
Experimental conditions |
|
32 |
|
|
items 4 and 7 |
|
|
|
4 |
Parameters to be measured |
|
42 |
|
|
item 5 |
|
|
|
5 |
Measuring methodologies |
|
38 |
|
|
item 3 |
|
|
|
6 |
Assay and model validation |
|
49 |
|
|
item 2 |
|
|
|
7 |
Modeling software |
|
42 |
|
|
item 9 |
|
|
|
8 |
Record keeping, data
coll/stor/acc |
|
15 |
|
|
item 1 |
|
|
|
9 |
Reporting and publishing |
|
17 |
|
|
item 8 |
|
|
|
10 |
Project management |
|
|
|
|
10 |
|
|
|
item 10 |
least frequently
mentioned |
|
|
|
A
more detailed description of the SOP items is given below |
|
|
1 |
Study object |
|
|
|
|
a. Type of cell or tissue |
|
|
|
|
|
|
|
|
|
b. Source, history,
description of cells |
|
|
|
|
|
c. Genome sequence |
|
|
|
|
|
|
d. Storage of stock
cultures |
|
|
|
|
|
|
|
|
|
e. Cultures freely
available |
|
|
|
|
|
2 |
Study design |
|
a. Exhaustive protocol |
|
|
|
|
|
|
b. Commonly available
standards/controls |
|
|
|
|
|
c. Appropriate statistics |
|
|
|
|
|
|
|
|
|
|
d. Regular SOP review and
authorization |
|
|
|
|
|
3 |
Experimental conditions |
|
|
|
a. Culture
medium/preparation/supplier |
|
|
|
|
|
|
b. Culture medium refresh
rate/method |
|
|
|
|
|
|
c. Culture medium pH
regulation |
|
|
|
|
|
|
d. Culture vessel
characteristics (roller bottle, spinner flask, fermentor) |
|
|
|
|
|
e. Oxygen supply
rate/method |
|
|
|
|
|
|
f. Temperature regulation |
|
|
|
|
|
|
|
g. Waste removal
rate/method |
|
|
|
|
|
|
h. Cells as monolayer, in
suspension, on carrier |
|
|
|
|
|
|
i. Shaking or stirring
rate/method |
|
|
|
|
|
|
j. Sampling rate/method |
|
|
|
|
|
|
|
k. Sample
identification/storage |
|
|
|
|
|
|
|
|
|
l. Perturbation
rate/method |
|
|
|
|
|
4 |
Parameters to be measured |
|
|
a. Metabolic indicators |
|
|
|
|
|
|
b. Enzyme
activity/specificity |
|
|
|
|
|
c.
Concentration/structure/size/folding of intracellular molecules |
|
|
|
|
|
d. Flux of
carbon/nitrogen/ions |
|
|
|
|
|
e. Cell density, size,
shape, motion, doubling time |
|
|
|
|
|
f. Spatial organization
of cell components |
|
|
|
|
|
g. Vesicle/follicle
formation |
|
|
|
|
|
h. Thermodynamics |
|
|
|
|
|
|
i. Membrane
permeability/fluidity/elasticity |
|
|
|
|
|
j. Secretory activity |
|
|
|
|
|
|
k. Electrical activity |
|
|
|
|
|
|
l. Genome expression |
|
|
|
|
|
|
m. Atom mapping |
|
|
|
|
|
|
|
|
|
n. Force-fields |
|
|
|
|
|
|
|
|
5 |
Measuring technologies |
|
|
|
a. Mass
(concentration/flux) |
|
|
|
|
|
b. Structure
(sequence/folding/shape) |
|
|
|
|
|
c. Size (Mol.Wt./Angstrom) |
|
|
|
|
|
d. Motion
(cilia/flagella/axon) |
|
|
|
|
|
e. Enzyme
(specificity/activity) |
|
|
|
|
|
f. Electrical phenomena
(charge, potential, polarity) |
|
|
|
|
|
|
|
|
|
g. Optical properties |
|
|
|
|
|
|
|
6 |
Assay and model
validation |
|
|
a. Independent
confirmation/validation of assays |
|
|
|
|
|
b. Assays to be specific,
precise, accurate, reproducible |
|
|
|
|
|
c. Iterative
confirmation/modification/validation of models |
|
|
|
|
|
d. Use models that can be
understood and interpreted |
|
|
|
|
|
|
|
|
|
e. Compare model outcome
to in vivo realities |
|
|
|
|
7 |
Modeling software |
|
|
a. Software published and
freely available |
|
|
|
|
|
b. Standardize
computation methods |
|
|
|
|
|
|
|
|
|
c. Guarantee reviewers'
access to modeling software |
|
|
|
|
8 |
Record keeping, data
coll/storage/access |
|
|
a. Lab notebooks to be
dated, authenticated, signed and countersigned |
|
|
|
|
b. Source and shelf-life
of all experimental compounds fully documented |
|
|
|
|
c. Electronic copy of
records maintained locally and centrally |
|
|
|
|
|
d. Unified storage method
of models in database |
|
|
|
|
|
|
|
|
|
e. Freele accessible
web-based databases |
|
|
|
|
|
9 |
Reporting/publishing |
|
|
a. Electronic template
for protocols, reports and articles |
|
|
|
|
|
c. Web-based European
Journal of Systems Biology |
|
|
|
|
|
d. Images, graphs tables
and legends to be included in full-text searches |
|
|
|
|
e. Selection procedure
for referees formalized |
|
|
|
|
|
|
|
|
|
f. Printed abstracts
available for public understanding, press and media |
|
|
10 |
Project management |
|
|
a.
Full description of objectives, timelines, deliverables, owners,
substitutes |
|
|
|
|
b. Regular team meetings
scheduled |
|
|
|
|
|
|
|
|
|
c. Acknowledgement of
funding and potential competing financial interest |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|