Statistical Genetics, May 14, 2020
This report from the Bits & Bäume conference 2018 in Berlin was first posted on ResearchGate as contribution to the GTTN (Global Timber Tracking Network)-project log in December 2018. The current version is slightly modified.
- The Bits & Bäume-Conference 2018
- Question 1: Hardware, energy use and general resource consumption
- Question 2: An integrated digital work environment and genomic profiling
- Question 3: Multispecies ethics
- Defining adequate applications of high tech and big data
The Bits & Bäume-Conference 2018
Last month, the Bits & Bäume (engl. Bits & Trees) conference on digitalization and sustainability in Berlin, Germany, brought together the, often disconnected, communities of environmentalists and digital human rights activists. The conference was organized and supported by a dozen+ well-known German NGOs, which are working nationally to internationally on environmental and conservation issues, social justice and digital rights. See the conference’s website at https://bits-und-baeume.org/en .
What I took home from the conference, was the view of the ‘Techies’ of how effective and successful the environmental movement is in explaining its issues, transporting them into society and thus reaching its goals, cp. pollution, recycling, nuclear power. On the other hand, the digital rights community suggested, even warned that as environmentalists we are quite naively using digital tools and are not careful enough with regard to our approach to information.
The subjects and questions raised at the meeting are in several ways directly of relevance to GTTN [the Global Timber Tracking Network-project]. Some of those concerns are very actively approached within GTTN. As for example, questions of database access to biodiversity information or of global social justice. Some issues, both on the sustainability, as well as, on the digitalization side, we as a network should be more aware of.
One of the areas that has questions associated, which need to be considered more closely, is population-level assignment of geographic origin using genetic data for forensic timber tracking and certification tools. Based on the discussions at the conference, I identified three questions, which I am presenting in the following.
Reviewing the available analytical methods for population assignment, applicable evolutionary theory and statistic basics on validation, I don’t see an alternative to genomic data that represent the genetic diversity across the distribution of a tree species. Only such genomic datasets will be sufficiently informative to robustly resolve patterns that arose in highly complex ecological and evolutionary systems. Such an approach is resource-expensive at across most of the inference process. It requires high-tech computing hardware and software and will aggregate, as well as, produce large amounts of data, information and insight.
Question 1: Hardware, energy use and general resource consumption
The hardware, energy use and general resource consumption for analyzing population-level genomic datasets are in themselves part of the reasons that natural forests are destroyed in the first place. The mining activities required for building computing hardware are directly deforesting landscapes. It seems not unusual that in many parts of the world, in addition, they are associated with illegal logging, which provides the financial means to factions fighting over the control of mines and regions rich in mineral resources.
Importantly, the demand for increasingly powerful computing infrastructures is embedded in the global North’s, which is GTTN’s origin, unsustainable resource consumption. An over-consumption that drives a well-understood vicious circle of deforestation, environmental degradation, social inequalities and political instabilization.
Therefore, what is the ecological and social sum total of high-tech genomics for biodiversity and forest preservation? Is a sustainable balance possible?
One answer can be that a highly differentiated and technologized-digitalized society, which, yet, manages to work on a long-term sustainable environmental footprint will be an essential prerequisite to ensure a positive overall balance of a resource-intense approach to conservation. In such a society, surplus resources can be available in some areas, which then can be utilized to fight by high-tech means the illegal degradation or exploitation of the environment.
However, currently, achieving the required balanced resource-use in the global North does not seem to be realistic any time soon.
Question 2: An integrated digital work environment and genomic profiling
A population-genomics approach will only be effective as a tool under real-world conditions with the development and implementation of an integrated digital work and inference environment. Such a statistical computing environment provides the data infrastructure and procedures for efficient data cleaning, exploration and data selection, as well as, the inference methodologies, predictive evaluations and hypothesis tests themselves. It will need to be open-source and community-based, not only for the size and complexity of the task, but most importantly for global social justice and equality. People, who are directly, nationally or regionally, most impacted by the overexploitation of local biodiversity and the destruction of natural forests, ecosystems and environments need to have open-access to the analytical conservation tools and the opportunity to develop the work environment further, adapting it to their needs.
However, as was already discussed within GTTN for the information stored in the database itself, eg. data on species occurrence, also the analytical environment can be (ab)used contrary to its original intentions. The statistical analysis routines will need to work effectively and efficiently for many tree species and any endangered wildlife, including marine fishes of commercial interest. These species generally have in common that they are long-lived, widespread and mostly sexually reproducing. In that, trees and fishes are not that different from humans. The inference environment, thus, will be applicable to humans, too. Today, genomics has become comparably inexpensive and its general bioinformatic pipelines are mainstream routines. Any power player, e. g. warlord or autocrat, or nondemocratic government can refine the existing open-access reference datasets on human genetic diversity for his, her or their purposes. The digital conservation-genomic work environment applied to such a reference dataset would provide a user-friendly and powerful tool for genomic profiling that has the potential to undermine basic human rights. Is this a possible cost we are willing to pay? How to deal with this?
On the other hand, even in democratic societies, it will be of importance that genomic profiling is based on open and transparent approaches. In that way, results can be questioned and checked. Reality is, that forensic genomic profiling is an area, in which envisioned and developing techniques, as well as, applications are moving ahead at a very rapid pace (see for example, Ennis 2017; Vidaki & Kayser 2017; Kayser & Parson 2018; Lillis et al. 2018; Molteni 2018).
Question 3: Multispecies ethics
Finally, the question arises, what kind of relationships to our co-creatures, the natural world and reality itself we are further ingraining and strengthening by implementing such an overarchingly technical approach to life and reality. Will we still be going into forests, enjoying nature and facing the wild places and their inhabitants? Or will we increasingly experience our surroundings second-hand digitally? Will we only see genomes and standing volume, instead of the woods? Will we still be able to build immediate, real-world relationships to our co-creatures that are respectful, considerate and aware of their independence, rights and own approach to life? Thus, will we be able to live a multispecies ethics (van Dooren et al. 2016)? Or will we force our point of view of rights and choices upon other species that share with us this planet?
This question might seem to be utterly farfetched and esoteric. However, it might be the most important of the three. Its answer is likely directly connected to the outlined potential answer to the first question. A multispecies ethics certainly will be most difficult to realize and live, starting out from the reality and point in history that we find us in.
Defining adequate applications of high tech and big data
These questions come at a time, when speakers close to the German Chaos Computer Club, a prominent hacker group internationally, at Bits & Bäume warned that we are at a critical phase of being swept away by a deluge of undemocratic, monopolized information use. A time, in which CRISP/CAS-editing of the human germline is made part of reality, disregarding ongoing doubts, critical reflections and evaluations. Raising questions with regard to, for example, if we really want and should do, what we are able to do. A time, when humanity does not seem to be sufficiently able to limit the anthropogenic rise in global temperature and calls for technological fixes (Borgmann 2012), as for example geoengineering, become more alluring.
I believe it is of importance to critical question, reflect on and discuss if the technical solutions that we plan to build and use, are part of the steps needed to actually solve humanity’s problems. If they are, we need to clarify under which conditions technical solutions will be adequate.
At Bits & Bäume, the question was formulated to the point: Is a technical approach part of the solution, or is it part of the problem?
Borgmann A. 2012. The setting of the scene: Technological fixes and the design of the good life. Pages 189-199 in Preston C (ed.) Engineering the Climate. Lexington Books, Lanham, MD USA.
Ennis C. 2017. Forensic DNA profiling might be about to take a big leap forward. Are we ready? Occam’s Corner. The Guardian, https://www.theguardian.com/science/occams-corner/2017/feb/06/forensic-dna-profiling-might-be-about-to-take-a-big-leap-forward-are-we-ready-epigenetics (accessed December 5, 2018).
Kayser M, Parson W. 2018. Transitioning from Forensic Genetics to Forensic Genomics. Genes (Basel) 9:3.
Lillis R, Kasler D, Chabria A, Stanton S. 2018. East Area rapist: GEDmatch provided DNA link to investigators. Sacto 911. The Sacramento Bee, https://www.sacbee.com/news/local/crime/article209987599.html (accessed December 7, 2018).
Molteni M. 2018. The creepy genetics behind the Golden State killer case. https://www.wired.com/story/detectives-cracked-the-golden-state-killer-case-using-genetics/ (accessed December 7, 2018).
van Dooren T, Kirksey E, Münster U. 2016. Mutlispecies Studies. Cultivating arts of attentiveness. Environmental Humanities 8:1-23.
Vidaki A, Kayser M. 2017. From forensic epigenetics to forensic epigenomics: broadening DNA investigative intelligence. Genome Biology 18:238.