Reservoir Computing with Random Chemical Systems

in ALIFE 2020: The 2020 Conference on Artificial Life (2020, July 14)

Top-down engineering of biomolecular circuits to perform specific computational tasks is notoriously hard and time-consuming. Current circuits have limited complexity and are brittle and application ... [more ▼]

Top-down engineering of biomolecular circuits to perform specific computational tasks is notoriously hard and time-consuming. Current circuits have limited complexity and are brittle and application-specific. Here we propose an alternative: we design and test a bottom-up constructed Reservoir Computer (RC) that uses random chemical circuits inspired by DNA strand displacement reactions. This RC has the potential to be implemented easily and trained for various tasks. We describe and simulate it by means of a Chemical Reaction Network (CRN) and evaluate its performance on three computational tasks: the Hamming distance and a short- as well as a long-term memory. Compared with the deoxyribozyme oscillator RC model simulated by Yahiro et al., our random chemical RC performs 75.5% better for the short-term and 67.2% better for the long-term memory task. Our model requires an 88.5% larger variety of chemical species, but it relies on random chemical circuits, which can be more easily realized and scaled up. Thus, our novel random chemical RC has the potential to simplify the way we build adaptive biomolecular circuits. [less ▲]

Gene selection for optimal prediction of cell position in tissues from single-cell transcriptomics

in Life Science Alliance (2020), 3(11), 202000867

Single-cell RNA-seq (scRNAseq) technologies are rapidly evolving. While very informative, in standard scRNAseq experiments the spatial organization of the cells in the tissue of origin is lost. Conversely ... [more ▼]

Single-cell RNA-seq (scRNAseq) technologies are rapidly evolving. While very informative, in standard scRNAseq experiments the spatial organization of the cells in the tissue of origin is lost. Conversely, spatial RNA-seq technologies designed to maintain cell localization have limited throughput and gene coverage. Mapping scRNAseq to genes with spatial information increases coverage while providing spatial location. However, methods to perform such mapping have not yet been benchmarked. To fill this gap, we organized the DREAM Single-Cell Transcriptomics challenge focused on the spatial reconstruction of cells from the Drosophila embryo from scRNAseq data, leveraging as silver standard, genes with in situ hybridization data from the Berkeley Drosophila Transcription Network Project reference atlas. The 34 participating teams used diverse algorithms for gene selection and location prediction, while being able to correctly localize clusters of cells. Selection of predictor genes was essential for this task. Predictor genes showed a relatively high expression entropy, high spatial clustering and included prominent developmental genes such as gap and pair-rule genes and tissue markers. Application of the Top-10 methods to a zebrafish embryo dataset yielded similar performance and statistical properties of the selected genes than in the Drosophila data. This suggests that methods developed in this challenge are able to extract generalizable properties of genes that are useful to accurately reconstruct the spatial arrangement of cells in tissues. [less ▲]

The Luxembourg Parkinson’s Study: A Comprehensive Approach for Stratification and Early Diagnosis

in Frontiers in Aging Neuroscience (2018), 10

While genetic advances have successfully defined part of the complexity in Parkinson’s disease (PD), the clinical characterization of phenotypes remains challenging. Therapeutic trials and cohort studies ... [more ▼]

While genetic advances have successfully defined part of the complexity in Parkinson’s disease (PD), the clinical characterization of phenotypes remains challenging. Therapeutic trials and cohort studies typically include patients with earlier disease stages and exclude comorbidities, thus ignoring a substantial part of the real-world PD population. To account for these limitations, we implemented the Luxembourg PD study as a comprehensive clinical, molecular and device-based approach including patients with typical PD and atypical parkinsonism, irrespective of their disease stage, age, comorbidities, or linguistic background. To provide a large, longitudinally followed, and deeply phenotyped set of patients and controls for clinical and fundamental research on PD, we implemented an open-source digital platform that can be harmonized with international PD cohort studies. Our interests also reflect Luxembourg-specific areas of PD research, including vision, gait, and cognition. This effort is flanked by comprehensive biosampling efforts assuring high quality and sustained availability of body liquids and tissue biopsies. We provide evidence for the feasibility of such a cohort program with deep phenotyping and high quality biosampling on parkinsonism in an environment with structural specificities and alert the international research community to our willingness to collaborate with other centers. The combination of advanced clinical phenotyping approaches including device-based assessment will create a comprehensive assessment of the disease and its variants, its interaction with comorbidities and its progression. We envision the Luxembourg Parkinson’s study as an important research platform for defining early diagnosis and progression markers that translate into stratified treatment approaches. [less ▲]

Fractalis: A scalable open-source service for platform-independent interactive visual analysis of biomedical data

in GigaScience (2018)

Background: Translational research platforms share the aim to promote a deeper understanding of stored data by providing visualization and analysis tools for data exploration and hypothesis generation ... [more ▼]

Background: Translational research platforms share the aim to promote a deeper understanding of stored data by providing visualization and analysis tools for data exploration and hypothesis generation. However, such tools are usually platform-bound and are not easily reusable by other systems. Furthermore, they rarely address access restriction issues when direct data transfer is not permitted. In this article we present an analytical service that works in tandem with a visualization library to address these problems. Findings: Using a combination of existing technologies and a platform-specific data abstraction layer we developed a service that is capable of providing existing web-based data warehouses and repositories with platform-independent visual analytical capabilities. The design of this service also allows for federated data analysis by eliminating the need to move the data directly to the researcher. Instead, all operations are based on statistics and interactive charts without direct access to the dataset. Conclusion: The software presented in this article has a potential to help translational researchers achieve a better understanding of a given dataset and quickly generate new hypothesis. Furthermore, it provides a framework that can be used to share and reuse explorative analysis tools within the community. [less ▲]

Device-based assessment through a mobile application in the Luxembourg Parkinson Cohort

in Basal Ganglia (2017, May), 8

Introduction: The project focuses on the integration device-based assessment (DBA) with a mobile application (mPower) into the longitudinal deeply-phenotyped HELP-PD (Health in the Elderly Luxembourgish ... [more ▼]

Introduction: The project focuses on the integration device-based assessment (DBA) with a mobile application (mPower) into the longitudinal deeply-phenotyped HELP-PD (Health in the Elderly Luxembourgish Population with a focus on Parkinson’s disease) cohort for patients with Parkinsonism in Luxembourg and the Greater Region to monitor frequency and degree of variation in symptoms of Parkinsonism, to identify potential sources and modulators of variation and to evaluate how symptoms are correlated with these modulators across patients. Methods: We integrate for the first time the mPower iOS app into a deeply phenotyped cohort. mPower is one of the first apps to use Apple’s Research Kit framework and combines a traditional survey-based approach with more granular and precise data gained from a person’s iPhone related to sensor- (e.g. step count, GPS-tracking) or task-based assessments (e.g. finger tapping, tremor detection, sustained phonation, simple gait analysis, memory test). Anonymized longitudinal data is sent to a repository, then retrieved, matched, and correlated with conventional HELP-PD data from a total of 47 screening instruments for motor and non-motor functions in Parkinsonism obtained from annual visits of study participants. 14 patients with clinically confirmed IPD are currently included in the pilot phase. Results/Discussion: We modified the mPower app and successfully integrated it into HELP-PD’s novel database infrastructure, allowing for a wide variety of analyses. The reporting system is able to handle multiple DBAs, with the implementation of an in-depth gait analysis system currently pending. Considerable attention was given to data protection. The system is currently fully functional with the pilot phase having started in June 2016. First correlations with traditional clinical data are planned for early 2017. [less ▲]

Configuration Symmetry and Performance Upper Bound of One-Dimensional Cellular Automata for the Leader Election Problem

in Journal of Cellular Automata (2015), 10(1-2), 1-21

Leader election plays a crucial role in numerous distributed protocols and biological societies. Yet, there is a fundamental gap in understanding its simplest variant employing components that are fully ... [more ▼]

Leader election plays a crucial role in numerous distributed protocols and biological societies. Yet, there is a fundamental gap in understanding its simplest variant employing components that are fully uniform, deterministic, and anonymous, such as in one-dimensional cellular automata (CA). In our previous work we found several binary CAs that elect a leader by using spatial-temporal patterns, domains and particles, which carry and exchange information across distance. The best strategy reached performance of 94 -99% with a modulo 6 limitation for the number of cells. As opposed to state-of-the-art distributed algorithms, a CA consists just of binary components without any extra memory or communication capabilities, and therefore operates with minimal resources possible. In this paper we identify fundamental limitations of leader election for one-dimensional CAs and formulate an upper bound on performance. We show that configurations that are symmetric or loosely-coupled are insolvable. The proportion of configurations that are insolvable decreases dramatically with the system size. Our findings could help to design more effective distributed protocols and also to model biological processes such as morphogenesis of cell differentiation, where leader election breaks symmetry in a newly formed organism. [less ▲]

Novel Methods for Learning and Adaptation in Chemical Reaction Networks

Doctoral thesis (2015)

State-of-the-art biochemical systems for medical applications and chemical computing are application-specific and cannot be re-programmed or trained once fabricated. The implementation of adaptive ... [more ▼]

State-of-the-art biochemical systems for medical applications and chemical computing are application-specific and cannot be re-programmed or trained once fabricated. The implementation of adaptive biochemical systems that would offer flexibility through programmability and autonomous adaptation faces major challenges because of the large number of required chemical species as well as the timing-sensitive feedback loops required for learning. Currently, biochemistry lacks a systems vision on how the user-level programming interface and abstraction with a subsequent translation to chemistry should look like. By developing adaptation in chemistry, we could replace multiple hard-wired systems with a single programmable template that can be (re)trained to match a desired input-output profile benefiting smart drug delivery, pattern recognition, and chemical computing. I aimed to address these challenges by proposing several approaches to learning and adaptation in Chemical Reaction Networks (CRNs), a type of simulated chemistry, where species are unstructured, i.e., they are identified by symbols rather than molecular structure, and their dynamics or concentration evolution are driven by reactions and reaction rates that follow mass-action and Michaelis-Menten kinetics. Several CRN and experimental DNA-based models of neural networks exist. However, these models successfully implement only the forward-pass, i.e., the input-weight integration part of a perceptron model. Learning is delegated to a non-chemical system that computes the weights before converting them to molecular concentrations. Autonomous learning, i.e., learning implemented fully inside chemistry has been absent from both theoretical and experimental research. The research in this thesis offers the first constructive evidence that learning in CRNs is, in fact, possible. I have introduced the original concept of a chemical binary perceptron that can learn all 14 linearly-separable logic functions and is robust to the perturbation of rate constants. That shows learning is universal and substrate-free. To simplify the model I later proposed and applied the ``asymmetric" chemical arithmetic providing a compact solution for representing negative numbers in chemistry. To tackle more difficult tasks and to serve more complicated biochemical applications, I introduced several key modular building blocks, each addressing certain aspects of chemical information processing and learning. These parts organically combined into gradually more complex systems. First, instead of simple static Boolean functions, I tackled analog time-series learning and signal processing by modeling an analog chemical perceptron. To store past input concentrations as a sliding window I implemented a chemical delay line, which feeds the values to the underlying chemical perceptron. That allows the system to learn, e.g., the linear moving-average and to some degree predict a highly nonlinear NARMA benchmark series. Another important contribution to the area of chemical learning, which I have helped to shape, is the composability of perceptrons into larger multi-compartment networks. Each compartment hosts a single chemical perceptron and compartments communicate with each other through a channel-mediated exchange of molecular species. Besides the feedforward pass, I implemented the chemical error backpropagation analogous to that of feedforward neural networks. Also, after applying mass-action kinetics for the catalytic reactions, I succeeded to systematically analyze the ODEs of my models and derive the closed exact and approximative formulas for both the input-weight integration and the weight update with a learning rate annealing. I proved mathematically that the formulas of certain chemical perceptrons equal the formal linear and sigmoid neurons, essentially bridging neural networks and adaptive CRNs. For all my models the basic methodology was to first design species and reactions, and then set the rate constants either "empirically" by hand, automatically by a standard genetic algorithm (GA), or analytically if possible. I performed all simulations in my COEL framework, which is the first cloud-based chemistry modeling tool, accessible at http://coel-sim.org. I minimized the amount of required molecular species and reactions to make wet chemical implementation possible. I applied an automatized mapping technique, Soloveichik's CRN-to-DNA-strand-displacement transformation, to the chemical linear perceptron and the manual signalling delay line and obtained their full DNA-strand specified implementations. As an alternative DNA-based substrate, I mapped these two models also to deoxyribozyme-mediated cleavage reactions reducing the size of the displacement variant to a third. Both DNA-based incarnations could directly serve as blue-prints for wet biochemicals. Besides an actual synthesis of my models and conducting an experiment in a biochemical laboratory, the most promising future work is to employ so-called reservoir computing (RC), which is a novel machine learning method based on recurrent neural networks. The RC approach is relevant because for time-series prediction it is clearly superior to classical recurrent networks. It can also be implemented in various ways, such as electrical circuits, physical systems, such as a colony of Escherichia Coli, and water. RC's loose structural assumptions therefore suggest that it could be expressed in a chemical form as well. This could further enhance the expressivity and capabilities of chemically-embedded learning. My chemical learning systems may have applications in the area of medical diagnosis and smart medication, e.g., concentration signal processing and monitoring, and the detection of harmful species, such as chemicals produced by cancer cells in a host (cancer miRNAs) or the detection of a severe event, defined as a linear or nonlinear temporal concentration pattern. My approach could replace “hard-coded” solutions and would allow to specify, train, and reuse chemical systems without redesigning them. With time-series integration, biochemical computers could keep a record of changing biological systems and act as diagnostic aids and tools in preventative and highly personalized medicine. [less ▲]

Training an asymmetric signal perceptron through reinforcement in an artificial chemistry

in Journal of the Royal Society, Interface (2014), 11(93),

State-of-the-art biochemical systems for medical applications and chemical computing are application-specific and cannot be reprogrammed or trained once fabricated. The implementation of adaptive ... [more ▼]

State-of-the-art biochemical systems for medical applications and chemical computing are application-specific and cannot be reprogrammed or trained once fabricated. The implementation of adaptive biochemical systems that would offer flexibility through programmability and autonomous adaptation faces major challenges because of the large number of required chemical species as well as the timing-sensitive feedback loops required for learning. In this paper, we begin addressing these challenges with a novel chemical perceptron that can solve all 14 linearly separable logic functions. The system performs asymmetric chemical arithmetic, learns through reinforcement and supports both Michaelis–Menten as well as mass-action kinetics. To enable cascading of the chemical perceptrons, we introduce thresholds that amplify the outputs. The simplicity of our model makes an actual wet implementation, in particular by DNA-strand displacement, possible. [less ▲]

COEL: A Web-based Chemistry Simulation Framework

in Stepney, Susan; Andrews, Paul (Eds.) CoSMoS 2014: Proceedings of the 7th Workshop on Complex Systems Modelling and Simulation (2014)

The chemical reaction network (CRN) is a widely used formalism to describe macroscopic behavior of chemical systems. Available tools for CRN modelling and simulation require local access, installation ... [more ▼]

The chemical reaction network (CRN) is a widely used formalism to describe macroscopic behavior of chemical systems. Available tools for CRN modelling and simulation require local access, installation, and often involve local file storage, which is susceptible to loss, lacks searchable structure, and does not support concurrency. Furthermore, simulations are often single-threaded, and user interfaces are non-trivial to use. Therefore there are significant hurdles to conducting efficient and collaborative chemical research. In this paper, we introduce a new enterprise chemistry simulation framework, COEL, which addresses these issues. COEL is the first web-based framework of its kind. A visually pleasing and intuitive user interface, simulations that run on a large computational grid, reliable database storage, and transactional services make COEL ideal for collaborative research and education. COEL's most prominent features include ODE-based simulations of chemical reaction networks and multicompartment reaction networks, with rich options for user interactions with those networks. COEL provides DNA-strand displacement transformations and visualization (and is to our knowledge the first CRN framework to do so), GA optimization of rate constants, expression validation, an application-wide plotting engine, and SBML/Octave/Matlab export. We also present an overview of the underlying software and technologies employed and describe the main architectural decisions driving our development. COEL is available at this http URL for selected research teams only. We plan to provide a part of COEL's functionality to the general public in the near future. [less ▲]

Complex Dynamics in Random DNA Strand Circuits

Poster (2013)