Functional Information: its potential for quantifying biodiversity and its relation to ecosystem functioning.

Abstract:

This highly interdisciplinary workshop unites knowledge and understanding from information theory, community ecology, and molecular and computational biology. It will be preceded by remotely managed collaboration; hence, its purpose is to consolidate developing ideas. We will explore the use of information theory to address the iDiv theme question of how biodiversity gives rise to ecological function. This will address the complexity of biological systems as dynamic networks, seeking to quantify systemic complexity and explore the potential for complexity metrics to improve understanding of how biodiversity contributes to ecological function. We will focus on “How to scale-up existing information-based measures from the molecular scale to all scales, including that of ecological networks”. The workshop shall examine information-theoretical methods to quantify biodiversity as complexity, in terms of information content, using network models of biological systems at multiple scales of organisation. We will explore ways to measure the contribution of biological complexity to ecosystem function by mapping the state-space of biological-network models in a range of ‘function’ metrics using a case study system to encourage its take-up by empirical scientists. Following the workshop we will apply our developments to estimate the functional information of the case study system and relate this to empirical measures of ecological function to test the value of this concept for practical prioritising of ecosystem attributes in conservation plans.


Who was meant to attended:
Christoph Adami †
 Mitchigan State University USA
 Genetics, complexity and theoretical biology
Larissa Albantakis
 University of Wisconsin USA
 Mathematical analysis of biological networks
Arto Annila
 University of Helsinki Finland
 Thermodynamcs of biology and complex systems
Tancredi Caruso †
 Queen’s University Belfast UK
 Community Ecology in soil systems
Christina Cobbold
 University of Glasgow UK
 Mathematics applied to biodiversity
Keith Farnsworth *
 Queen’s University Belfast UK Theoretical Biology: Information and life.
Ivo Große
 Martin-Luther-University Germany
 Bioinformatics and Biodiversity
Ute Jacob †
 University of Hamburg Germany
 Functional Biodiversity of marine systems
Olga Lyashevska
 Royal Netherlands Institute for Sea Research NL
 Biodiversity quantification / Biostatistics
Timothee Poisot †
 University of Canterbury, New Zealand
 Community Ecology / modelling / BEF relationships
Axel Rossberg
 Centre for Ecology Fisheries Aquatic Science. UK
 Theoretical Community Ecology
Tom Schneider
 National Cancer Institute, Maryland USA.
 Information theory for molecular biology
Sara Imari Walker † Arizona State University USA
 Informational approach to origin of life

* Coordinator
† Could not attend (but as you see below, some sent substitutes)




The 'A' team: a happy group photo of sFIND at the German Centre for Integrative Biodiversity Research (iDiv) in Leipzig.
group photo
From left to right (as you see them)
back row - Tom Schneider, Axel Rossberg, Keith Baverstock, Hyunju Kim, Larissa Albantakis, Christina Cobbold, Arto Annila, Ivo Grosse
front row - Keith Farnsworth, Olga Lyashevska, David Ardell, Lars Gamfeldt.


Aims and scope of the workshop

The overall aim is to explore the prospect for the understanding of biodiversity and its role for ecological function, using the information theory currently developed for molecular biology. The focus of the workshop will be examining how we might scale the functional information concepts from molecular to ecological levels of organisation. Its specific aim is to seek a practical measure of FI for a biological system, which we describe generally as a network of interactions. This will provide an important link in the chain that leads from quantifying biodiversity to predicting the resulting ecological function.

The workshop will seek to develop a rigorous definition of FI in biological networks and identify the quantifiable characteristics appearing at the molecular scale that can be used at higher network scales. This aim is centred on the overarching iDiv question “How does biodiversity emerge (‘emergence’)?” Since emergence is a property of complex systems and life may be thought of as organised complexity, from which phenomena emerge at several organisational scales to form a nested hierarchy of pattern, the emergence of biodiversity is a consequence of the functional information embodied by this complexity. The idea is that by defining these concepts with mathematical rigour, we may quantify ‘functional’ biodiversity over all scales of biological organisation. The practical outcome will be a general mathematical framework, with appropriate metrics, for describing BEF relations of biological systems. It will partition contributions of functional information from the individual level, and distinctly, from the network complexity.

This work will be recorded as a foundational paper, authored by workshop participants. Its conclusions will directly address the second iDiv overarching question “What are the consequences of biodiversity for the functioning of ecosystems (‘consequences’)?”.

Scientific Objectives

We will examine whether the concept of biodiversity as functional information clarifies and generalises BEF relations, particularly in relation to ecological network characteristics.
In doing so we intend to demonstrate the implications of our analysis for the following questions:

Are results from diversity-function experiments valid at larger scales (space and time)?
We are looking for general characteristics of biological networks that embody functionality  at an extended range of scales. Interpreting FI as a scale-independent metric for biodiversity will enable general predictions about BEF. In particular we will investigate whether new sources of FI emerge at larger scales of organisation,, especially those normally missed by BEF experiments.

How can we optimize conservation strategies to safeguard the full range of biological information?
Through quantifying biological information, not only within, but among the components of a functioning ecosystem (the network information), we can show the dependence of ecosystem function on the FI embodied at the system level. This plugs a knowledge gap, enabling conservation managers to take account of systemic complexity, in addition to the individual organisms.

How and why are different types of diversities (genetic, species, functional, interaction) correlated?
Our aim is to unite these by generalising all forms of diversity as cases of functional information. That will enable a quantitative understanding of their inter-relationships.
Specifically we shall pursue the following scientific objectives (1 and 2 in the workshop, based on preparatory work, 3 and 4 following it as outcomes:

Objective 1.
Establish a rigorous definition of functional information for biological networks, spanning molecular biology and ecology, identifying the common features among these different scales of biological organisation (note this may develop throughout the workshop).

Objective 2.
Identify the quantifiable characteristics of functional information appearing at the scale of molecular networks, and find ways to  apply these at the scale of ecological networks.

Objective 3.
From these quantifiable characteristics, estimate the FI of an ecological network, using worked examples from soil and marine systems as an illustration.

Objective 4.
As a follow-up, use ecological network scale FI to quantify BEF relations in systematic disassembly simulation experiments, modelled on case-study ecological systems. This objective will be realised through follow-up research projects.


Work plan

BEFORE:
1) Establishment of online research collaboration, with forum and document sharing and agreed scope and overall aims, consistent with the workshop aims and objectives.
2) Review of conceptual and quantitative measures of functional information and diversity at molecular, organism and ecological levels of biological organisation.

3) Selection and description of worked example from soil and marine communities.
4) Agreed plan for multi-disciplinary publication: a synthesis explaining how the concept of functional information is useful in understanding biodiversity and BEF relationships over a range of scales of biological organisation.

DURING:
5) The workshop (see time-table of events below).

AFTER:
6) Follow-up completion and submission of the multi-author paper and applications for PhD studentships and further grant funding for collaborative research among the group, especially targeting the development of worked examples that use the concepts and methods we develop in the multi-author paper.
The workshop will consist of a presentation-discussion-writing cycle, for which participants will have prepared their subject-specific contributions in advance. Participants will synthesise their prepared contributions to form a synthesis document. This document will be the basis for a major cross-discipline paper that states the new interpretation of biodiversity as the biological embodiment of functional information, where the function is ecological function, from which the BEF relationship follows. The workshop will use a case-study ecosystem to illustrate the power of the new synthesis for quantifying biodiversity and its resulting ecological function. This case-study will be supplied by participants with system-specific expertise in soil and marine community structure.
The working group will start preparing an outline of the planed publication from the beginning of 2015. This outline will be presented and discussed at the workshop. Preparation documents and discussions will be supported by this website (http://www.whatlifeis.info).
After the workshop, the indicated deliverables will be prepared. First drafts of manuscripts are expected by March 2016, submission is anticipated for July 2016.

Provisional Timetable
Time
Led by* Activity
Monday a.m.
 KF / IG
 Introduction and setting the scene:
from molecular to ecological information; in which each participant gives a brief introduction to their perspective.
p.m.
 CC / OL/ KF
 What really is biodiversity?
Tuesday a.m.
 CA/ IG
 Molecular measures of complexity and function
p.m.
 LA/ CA
 Quantifying information characteristics of biological networks
Wed.  a.m.
 TS
 Self-organisation and Shannon applied to biological systems.
p.m.
 SW/ AA/ CA
 Information, Entropy and the role of complexity in sustaining life
Thursday a.m.
 AR / TP
 How to scale information measures to ecosystem scale
p.m.
 TC /UJ
 Real-life consequences of ecological complexity
Friday a.m.
 AA / TS/ SW
 Integrating metrics of diversity from molecules to ecosystems
p.m.
 KF/ IG
 A practical and concrete plan for publication

* Led by does not mean they do all the work, rather it means they are the responsible coordinator for the session.

Here are a few (references to) papers that are well worth looking at in preparation:

Adami, C. Information-Theoretic Considerations Concerning the Origin of Life. (2014).  (from arXiv:1409.0590v1).

Lizier, J. T. (2014). JIDT: An information-theoretic toolkit for studying the dynamics of complex systems. Frontiers in Robotics and AI 1:11 (from arXiv:1408.3270v2). 

Lyashevska, O. and Farnsworth, K. D. (2012). How many dimensions of biodiversity do we need? Ecological Indicators. 18:485–492.

Schneider, T.D. (2010). 70% efficiency of bistate molecular machines explained by information theory, high dimensional geometry and evolutionary convergence. Nucleic Acids Res. 38, 5995-6006.

Walker, S.I. Top-Down Causation and the Rise of Information in the Emergence of Life. (2014). Information. 5, 424-439.

OUTCOME
The result of this meeting was a commitment to write at least one paper. I (Keith) had hoped there would be more, but at least we did manage to combine the efforts of Keith, Larissa and Tancredi to write a paper on Biological Function: Farnsworth, K.D.; Albantakis, L.; Caruso, T. (2017). Unifying concepts of biological function from molecules to ecosystems. Oikos, doi: 10.1111/oik.04171.

We have kept in touch with the German Synthesis Centre for Biodiversity and are working on a plan to develop the use of information metrics (such as the Phi in  Integrated Information Theory) to characterise the contribution of community organisation towards biodiversity in terms of function (the BEF attributable to system organisation).