Agnostic Way Research Articles

Computing Random Forest-distances in the presence of missing data

In this article, we study the problem of computing Random Forest-distances in the presence of missing data. We present a general framework which avoids pre-imputation and uses in an agnostic way the information contained in the input points. We centre our investigation on RatioRF, an RF-based distance recently introduced in the context of clustering and shown to outperform most known RF-based distance measures. We also show that the same framework can be applied to several other state-of-the-art RF-based measures and provide their extensions to the missing data case. We provide significant empirical evidence of the effectiveness of the proposed framework, showing extensive experiments with RatioRF on 15 datasets. Finally, we also positively compare our method with many alternative literature distances, which can be computed with missing values.

Abstract 3097: CETSA HT: Employing rapid high-throughput target engagement technology in primary screening of P53

Abstract The confirmation of target engagement for candidate drugs is a fundamental requisite in drug discovery. A significant portion of projects fail to reach the clinic due to lack of efficacy or failure to show lead candidates interacting with the intended target in a more complex environment. The utility of using the CETSA® (CEllular Thermal Shift Assay) technology to measure target engagement in intact cells is increasingly exploited in high throughput screening, where examples demonstrate low false positive rates in screens of up to 500K compounds. The p53 tumor suppressor protein, a crucial guardian against cancer, functions as a transcription factor that mediates cancer suppression by regulating genes involved in a wide range of cellular processes such as DNA repair, cell cycle and apoptosis. Unfortunately, frequent mutations in the TP53 gene hinder p53's tumor suppressor role in human cancers, thus designating it as a promising target for anticancer therapy. However, attempts to restore p53 function for therapeutic purposes have not yet advanced to clinical approval, highlighting the complexities in leveraging p53 for cancer treatment. Its undruggable nature, attributed to the absence of accessible deep pockets, lack of enzymatic activity, and its intracellular location, further complicates targeted interventions with both low and high molecular weight drugs. Here, we demonstrate the analysis of a high-throughput plate based CETSA® HT screen on the challenging target p53, performed using AlphaLISA® SureFire® Ultra™ technology and our newly developed high-throughput screening platform. Hits identified from this screen are privileged from the onset by binding to the target in a physiologically relevant cellular matrix and in a binding-mode agnostic way. This is in contrast to traditional screening methods targeting activity or in vitro interaction where cellular activity is addressed in later stages. This assay cascade highlights a rapid target-to-hit lead generation strategy placing the emphasis on finding chemical matter that interacts with endogenous target in living cells first, followed by analysis of hit material activity. The ‘shortcut’ to chemical matter active in living cells represents significant potential cost savings in comparison to assay development time on recombinantly produced proteins and/or modified cell lines for traditional HTS screening approaches that still rely on target engagement assessment downstream. Citation Format: Merve Kacal, Tomas Friman, Laurence Arnold, Victoria Brehmer, Daniel Martinez Molina, Stina Lundgren. CETSA HT: Employing rapid high-throughput target engagement technology in primary screening of P53 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3097.

Abstract 3096: CETSA Navigate HT: A new paradigm for high-throughput screening

Abstract The CEllular Thermal Shift Assay (CETSA) technology is a powerful way to assess target engagement across various matrixes. By the introduction of the high-throughput format (CETSA Navigate HT), this technology offers a reliable high-throughput screening method in natural cellular environments that can successfully be applied in compound screening for hit identification and confirmation already during lead generation. Due to its broad applicability to native, full length protein targets in the live cellular environment, without the need for expensive and time-consuming protein and/or compound labelling efforts, CETSA Navigate HT may be regarded as a new paradigm for high-throughput screening and actives generation. As a case study to demonstrate the applicability of CETSA Navigate HT in hit identification, we have applied the technology for primary screening of a 12000 compound library for CDK4, a protein targeted in several cancers. Hits identified from this screen are privileged from the onset by binding to the target in a physiologically relevant cellular matrix and in a binding-mode agnostic way. This is in contrast to traditional screening methods targeting activity or in vitro interactions where cellular activity is addressed in later stages. Data from follow up studies using functional assays will be presented as well as mass spectrometry based CETSA selectivity profiling and comparison to commercially available de novo CDK4/6 inhibitors. Citation Format: Helena Almqvist, Victoria Brehmer, Laurence Arnold, Tomas Friman, Stina Lundgren, Daniel Martinez Molina. CETSA Navigate HT: A new paradigm for high-throughput screening [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3096.

DNA mismatch repair deficient cancer - Emerging biomarkers of resistance to immune checkpoint inhibition.

The DNA mismatch repair pathway is involved in the identification, excision, and repair of base-base mismatches and indel loops in the genome. Mismatch repair deficiency occurs in approximately 20% of all cancers and results in a type of DNA damage called microsatellite instability. In 2017, the immune checkpoint inhibitor, Pembrolizumab, an anti-PD-1 therapy, was approved for use in all unresectable or metastatic tumours that were mismatch repair deficient or had high microsatellite instability regardless of tissue origin. This landmark approval was the first time a drug had been approved in a site agnostic way, but accumulating data has revealed that up to 50% of mismatch repair deficient tumours are refractory to treatment and there is a huge amount of variability in the therapeutic benefit amongst responders. Several mechanisms of resistance to immune checkpoint blockade for mismatch repair deficient cancers have been identified but our understanding of what is driving resistance in a proportion of patients remains lacking. In this review article, we discuss the emerging mechanisms of resistance which may enable optimal stratification of patients for treatment with immune checkpoint inhibitors in the future.

A Unified Framework for Manipulating N-dimensional Astronomical Data and Coordinate Transformations in Python: The NDCube 2 and Astropy APE-14 World Coordinate System APIs

The NDCube 2 API is a Python application programming interface (API) for storing and manipulating N-dimensional coordinate-aware astronomical data. While there are Python packages for handling astronomical data and coordinate transformations separately and for handling specific combinations of dimensions and transformations, none provide a unified and agnostic way of handling them simultaneously. This leads to a proliferation of different APIs for conducting the same analysis tasks on similar types of observations and introduces technical barriers between multi-instrument studies and cross-community collaboration. In this paper, we outline how the NDCube 2 API and its implementation in the open-source, community-developed ndcube package, together with the AstroPy WCS API, help to solve this problem. We discuss the guiding principles underpinning the API design and provide examples of how it is already being used to serve broad sections of the astronomy community, including agency-funded missions. The aim of this paper is to help users better understand the purpose and potential of the NDCube 2 API and ndcube package and hence how to more effectively deploy them in scientific analyses and software development.

Enabling 'AI'? The situated production of commensurabilities.

How can we examine so-called 'artificial intelligence' ('AI') without turning our backs on the STS tradition that questions both notions of artificiality and intelligence? This special issue attempts a step to the side: Instead of considering 'AI' as something that does or does not exist (and then taking a position on its benefits or harms), its ambition is to document, in an empirical and agnostic way, the performances that make, sometimes, 'AI' appear or disappear in situation. And it comes out, from this perspective, that 'AI' could be considered a vast commensuration undertaking.

Dual ICA to extract interacting sets of genes and conditions from transcriptomic data.

One of the challenges in RNA-Seq studies is finding subsets of genes that share a common mechanism of action or are associated with a regulon/pathway. Existing approaches often extract modules that reflect quantitative similarities (such as genes with correlated log-fold-changes) but do not adequately capture biological significance. In this work, we propose the Dual ICA methodology, which provides an agnostic way to extract "interacting modules" composed of sets of genes and conditions that exhibit strong associations. Dual ICA involves performing Independent Component Analysis (ICA) twice, once on the genes and once on the conditions. Using the resulting signal matrices, we extract respective sets of genes and conditions. The interaction between these sets is quantified using the coefficients from a linear regression and significance is determined through the Wald test and Z-score filtering. These coefficients are equivalent to the outer product of independent components obtained from the two signal matrices. Not only do the gene sets extracted align with known regulons, but the significant interacting modules they instantiate also encompass conditions that influence the expression of these regulons through shared mechanisms of action. Compared to traditional unsupervised clustering methods, Dual ICA demonstrates superior performance and provides explicit gene-condition sets for exploring functional relationships.

An Adaptable and Agnostic Flow Scheduling Approach for Data Center Networks

Cloud applications have reshaped the model of services and infrastructure of the Internet. Search engines, social networks, content delivery and retail and e-commerce sites belong to this group of applications. An important element in the architecture of data centers where these applications run is the communication infrastructure, commonly known as data center networks (DCNs). A critical challenge DCNs have to address is the processing of the traffic of cloud applications, which due to its properties is essentially different to the traffic of other Internet applications. In order to improve the responsiveness and throughput of applications, DCNs should be able to prioritize short flows (a few kB) over long flows (several MB). However, given the time and space variations the traffic presents, the information about flow sizes is not available in advance in order to plan the flow scheduling. There has been a wealth of solutions developed in this space, and prior work includes flow scheduling mechanisms optimizing for a specific workload but fall short when workloads are not known in advance, or comprise a collection of applications changing dynamically. In this paper, we present an adaptable mechanism called Adaptable Workload-Agnostic Flow Scheduling (AWAFS). It is an adaptable approach that can adjust in an agnostic way the scheduling configuration of DCN forwarding devices. This agnostic adjustment contributes to reduce the Flow Completion Time (FCT) of those short flows, representing around 85% of the traffic handled by cloud applications. AWAFS operates by observing the traffic and detecting statistical properties that provide a hint to adapt the scheduling parameters. Our evaluation results based on simulation show that AWAFS can reduce the average FCT of short flows between 16.9% and 45.2% when compared to the best existing agnostic non-adaptable solution, without inducing starvation on long flows. Indeed, it can provide improvements as high as 39% for long flows. Additionally, AWAFS can improve the FCT for short flows in scenarios with high heterogeneity in the traffic present in the network, with a reduction up to 5% for the average FCT and 15% for the tail FCT.

Suspect and non-targeted screening-based human biomonitoring identified 74 biomarkers of exposure in urine of Slovenian children

Human exposure to organic contaminants is widespread. Many of these contaminants show adverse health effects on human population. Human biomonitoring (HBM) follows the levels and the distribution of biomarkers of exposure (BoE), but it is usually done in a targeted manner. Suspect and non-targeted screening (SS/NTS) tend to find BoE in an agnostic way, without preselection of compounds, and include finding evidence of exposure to predicted, unpredicted known and unknown chemicals.This study describes the application of high-resolution mass spectrometry (HRMS)-based SS/NTS workflow for revealing organic contaminants in urine of a cohort of 200 children from Slovenia, aged 6–9 years. The children originated from two regions, urban and rural, and the latter were sampled in two time periods, summer and winter. We tentatively identified 74 BoE at the confidence levels of 2 and 3. These BoE belong to several classes of pharmaceuticals, personal care products, plasticizers and plastic related products, volatile organic compounds, nicotine, caffeine and pesticides. The risk of three pesticides, atrazine, amitraz and diazinon is of particular concern since their use was limited in the EU. Among BoE we tentatively identified compounds that have not yet been monitored in HBM schemes and demonstrate limited exposure data, such as bisphenol G, polyethylene glycols and their ethers. Furthermore, 7 compounds with unknown use and sources of exposure were tentatively identified, either indicating the entry of new chemicals into the market, or their metabolites and transformation products. Interestingly, several BoE showed location and time dependency.Globally, this study presents high-throughput approach to SS/NTS for HBM. The results shed a light on the exposure of Slovenian children and raise questions on potential adverse health effects of such mixtures on this vulnerable population.

Cyc-DEP: Cyclic immunofluorescence profiling of particles collected using dielectrophoresis.

Cancer is a highly heterogenous disease that requires precise detection tools and active surveillance methods. Liquid biopsy assays provide an agnostic way to follow the complex trajectory of cancer, providing better patient stratification tools for optimized treatment. Here, we present the development of a low-volume liquid biopsy assay called cyc-DEP (cyclic immunofluorescent imaging on dielectrophoretic chip) to profile biomarkers collected on a dielectrophoretic microfluidic chip platform. To enable on-chip cyclic imaging, we optimized a fluorophore quenching method and sequential rounds of on-chip staining with fluorescently conjugated primary antibodies. cyc-DEP allows for the quantification of a multiplex array of proteins using 25µl of a patient plasma sample. We utilized nanoparticles from a prostate adenocarcinoma (LNCaP) cell line and a panel of six target proteins to develop our proof-of-concept technique. We then used cyc-DEP to quantify blood plasma levels of target proteins from healthy individuals, low-grade and high-grade prostate cancer patients (n=3 each) in order to demonstrate that our platform is suitable for liquid biopsy analysis in its present form. To ensure accurate quantification of signal intensities and comparisons between different samples, we incorporated a signal intensity normalization method (fluorescent beads) and a custom signal intensity quantification algorithm that account for the distribution of signal across hundreds of collection regions on each chip. Our technique enabled a threefold improvement in multiplicity for detecting proteins associated with fluid samples, opening doors for early detection, and active surveillance through quantification of a multiplex array of biomarkers from low-volume liquid biopsies.

Abstract 5213: Genetically modified Tie-2 expressing monocytes target IFN-α2 to the glioblastoma tumor microenvironment (TME): Preliminary data from the TEM-GBM Phase 1/2a study

Abstract Increasing clinical use of immune checkpoint inhibitors testifies to the importance of modulating the immune TME to obtain meaningful anti-tumor immune responses. Acting only on T lymphocytes may, however, not be sufficient, e.g. in immunologically-cold tumors or due to de novo or acquired resistance. Moreover, immune-related AEs remain hurdles of T cell therapies. To overcome these limitations and to awaken the immune system in an agnostic way against the tumor, we have developed a genetically modified cell-based autologous hematopoietic stem cell platform (Temferon) delivering immunotherapeutic payloads into the TME through Tie-2 expressing monocytes (TEMs), a subset of tumor infiltrating macrophages. TEM-GBM is an ongoing open-label, Phase 1/2a dose-escalating study evaluating the safety & efficacy of Temferon in up to 21 newly diagnosed patients with glioblastoma & unmethylated MGMT promoter assigned to 7 different cohorts (3 pts each) differing by Temferon dose (0.5-4.0x106/kg) and conditioning regimen (BCNU+ or Busulfan+Thiotepa). By Oct 15th, 2021, 15 pts (cohort 1-5) had received escalating doses of Temferon with a median follow up of 267 days (range: 60-749). Rapid engraftment and hematological recovery from nonmyeloablative conditioning occurred in all pts. Temferon-derived differentiated cells, as determined by the presence of vector genomes in the DNA, were found at increasing proportions in PB and BM, reaching up to 30% at 1 month for the highest cohorts tested (2.0x106/kg) and persisting up to 18 months, albeit at lower levels. Despite the significant proportion of engineered cells, only very low median concentrations of IFNα were detected in the plasma (D+30, 5.9; D+90, 8.8pg/mL) and in the cerebrospinal fluid (D+30, 1.5; D+90, 2.4pg/mL), indicating tight regulation of vector expression. SAEs were mostly attributed to conditioning chemotherapy (e.g. infections) or disease progression (e.g. seizures). 1 SUSAR (persistent GGT elevation) has occurred. Median OS is 14 mth from surgery (11 mth post Temferon). Four pts from the low dose cohorts underwent 2nd surgery. These recurrent tumors contained gene-marked cells and expressed IFN-responsive genes, indicative of local IFNα release by TEMs. In 1 pt, a stable lesion (as defined by MRI) had a higher proportion of T cells & TEMs, an increased IFN-response signature and myeloid re-programming revealed by scRNAseq, as compared to a synchronous, progressing tumor. TCR sequencing of blood and tumor samples showed a post-treatment increase in the cumulative frequency of tumor-associated T cell clones identified in 1st and 2nd surgery specimens (up to 4 out of 9 subjects). These results provide initial evidence for on-target activity of Temferon in GBM, to be consolidated with longer follow up in the higher dose cohorts. Citation Format: Bernhard Gentner, Gaetano Finocchiaro, Francesca Farina, Marica Eoli, Alessia Capotondo, Elena Anghileri, Matteo Barcella, Maria Grazia Bruzzone, Matteo Giovanni Carrabba, Valeria Cuccarini, Giorgio D'Alessandris, Francesco Di Meco, Valeria Ferla, Paolo Ferroli, Filippo Gagliardi, Federico Legnani, Pietro Mortini, Matteo Maria Naldini, Alessandro Olivi, Roberto Pallini, Monica Patanè, Rosina Paterra, Bianca Pollo, Marco Saini, Silvia Snider, Valentina Brambilla, Stefania Mazzoleni, Andrew Zambanini, Carlo Russo, Luigi Naldini, Fabio Ciceri. Genetically modified Tie-2 expressing monocytes target IFN-α2 to the glioblastoma tumor microenvironment (TME): Preliminary data from the TEM-GBM Phase 1/2a study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 5213.

Harnessing genetically engineered hematopoietic progenitor cells to redirect the tumor immune microenvironment against glioblastoma (TEM-GBM Study).

2040 Background: Immunotherapies represent powerful tools that are transforming the treatment of many cancers. However, immune dysfunction in cancer is multifactorial requiring multiple points of action, especially in immunologically-cold tumors. Methods: We have developed a genetically modified, autologous hematopoietic stem cell-based platform designed to deliver Interferon-alpha (IFNa) specifically into the tumor microenvironment through Tie-2 expressing monocytes (Temferon), in order to activate the immune system in an agnostic way against the tumor and re-establish immunosurveillance. Results: As of Jan 2022, 3 escalating doses of Temferon (from 0.5 to 2.0x106/kg) were tested across 15 patients assigned to 5 cohorts affected by newly diagnosed, unmethylated MGMT glioblastoma (GBM). The follow-up range from surgery is 5 – 27 mo (3 – 24 mo after Temferon). In all patients, we observed rapid engraftment of gene modified progenitors and fast recovery from sub-myeloablative conditioning (median engraftment across all the cohorts: Neu D+13, PLT D+14). Temferon-derived differentiated cells, as determined by the presence of vector genomes in the DNA, were found at increasing proportions in blood and bone marrow, reaching up to 30% at 1 mo for the highest dose cohorts tested and persisting up to 18 mo, albeit at lower levels. Despite the significant proportion of engineered cells, only very low median concentrations of IFNα were detected in the plasma (D+30, 5.9; D+90, 8.8pg/mL) and in the CSF (D+30, 1.5; D+90, 2.4pg/mL), indicating tight regulation of vector expression. SAEs were mostly attributed to conditioning chemotherapy (e.g. infections) or disease progression (e.g. seizures). 1 SUSAR (persistent GGT elevation) has occurred. Median OS is 14 mo from surgery (10 mo post Temferon). A patient from cohort 3, had at D+120 disease progression with two distant enhancing lesions, and increased tumor necrosis. One year following Temferon, with no 2nd line therapy added, there was approximately 40% reduction in enhancing tumor volume compared to D+180. Four pts from the low dose cohorts underwent 2nd surgery. Vector genomes were detectable in tumor biopsies. Single cell RNA seq performed on CD45+ cells purified from the GBM TME highlighted the presence of an Interferon gene signature in all patients, resulting in macrophage repolarization in some of them. Conclusions: Our interim results show that Temferon is well tolerated, with no dose limiting toxicities identified to date. The results provide initial evidence of Temferon’s potential to modulate the TME of GBM patients. Clinical trial information: NCT03866109.

Vision based crown loss estimation for individual trees with remote aerial robots

With the capability of capturing high-resolution imagery data and the ease of accessing remote areas, aerial robots are becoming increasingly popular for forest health monitoring applications. For example, forestry tasks such as field surveys and foliar sampling which are generally manual and labour intensive can be automated with remotely controlled aerial robots. In this study, we propose two new online frameworks to quantify and rank the severity of individual tree crown loss. The real-time crown loss estimation (RTCLE) model localises and classifies individual trees into their respective crown loss percentage bins. Experiments are conducted to investigate if synthetically generated tree images can be used to train the RTCLE model as real images with diverse viewpoints are generally expensive to collect. Results have shown that synthetic data training helps to achieve a satisfactory baseline mean average precision (mAP) which can be further improved with just some additional real imagery data. We showed that the mAP can be increased approximately from 60% to 78% by mixing the real dataset with the generated synthetic data. For individual tree crown loss ranking, a two-step crown loss ranking (TSCLR) framework is developed to handle the inconsistently labelled crown loss data. The TSCLR framework detects individual trees before ranking them based on some relative crown loss severity measures. The tree detection model is trained with the combined dataset used in the RTCLE model training where we achieved an mAP of approximately 95% suggesting that the model generalises well to unseen datasets. The relative crown loss severity of each tree is estimated, with deep representation learning, by a probabilistic encoder from a fully trained variational autoencoder (VAE) model. The VAE is trained end-to-end to reconstruct tree images in a background agnostic way. Based on a conservative evaluation, the estimated crown loss severity from the probabilistic encoder generally showed moderate agreement with the expert’s estimation across all species of trees present in the dataset. All the software pipelines, the dataset, and the synthetic dataset generation can be found in the GitHub link.

Performance evaluation of data-driven techniques for the softwarized and agnostic management of an N×N photonic switch

The emerging Software Defined Networking (SDN) paradigm paves the way for flexible and automatized management at each layer. The SDN-enabled optical network requires each network element’s software abstraction to enable complete control by the centralized network controller. Nowadays, silicon photonics due to its low energy consumption, low latency, and small footprint is a promising technology for implementing photonic switching topologies, enabling transparent lightpath routing in re-configurable add-drop multiplexers. To this aim, a model for the complete management of photonic switching systems’ control states is fundamental for network control. Typically, photonics-based switches are structured by exploiting the modern technology of Photonic Integrated Circuit (PIC) that enables complex elementary cell structures to be driven individually. Thus PIC switches’ control states are combinations of a large set of elementary controls, and their definition is a challenging task. In this scenario, we propose the use of several data-driven techniques based on Machine Learning (ML) to model the control states of a PIC N×N photonic switch in a completely blind manner. The proposed ML-based techniques are trained and tested in a completely topological and technological agnostic way, and we envision their application in a real-time control plane. The proposed techniques’ scalability and accuracy are validated by considering three different switching topologies: the Honey-Comb Rearrangeable Optical Switch (HCROS), Spanke-Beneš, and the Beneš network. Excellent results in terms of predicting the control states are achieved for all of the considered topologies.

Separating astrophysics and geometry in black hole images

The observation of the shadow of the supermassive black hole $\mathrm{M}{87}^{*}$ by the Event Horizon Telescope (EHT) is sensitive to the spacetime geometry near the circular photon orbit and beyond, and it thus has the potential to test general relativity in the strong field regime. Obstacles to this program, however, include degeneracies between putative deviations from general relativity and both the description of the accretion flow and the uncertainties on ``calibration parameters,'' such as, e.g., the mass and spin of the black hole. In this work, we introduce a formalism, based on a principal component analysis, capable of reconstructing the black hole metric (i.e., the ``signal'') in an agnostic way, while subtracting the ``foreground'' due to the uncertainties in the calibration parameters and the modeling of the accretion flow. We apply our technique to simulated mock data for spherically symmetric black holes surrounded by a thick accretion disk. We show that separation of signal and foreground may be possible with next generation EHT-like experiments.

Via machinae: Searching for stellar streams using unsupervised machine learning

ABSTRACT We develop a new machine learning algorithm, via machinae, to identify cold stellar streams in data from the Gaia telescope. via machinae is based on ANODE, a general method that uses conditional density estimation and sideband interpolation to detect local overdensities in the data in a model agnostic way. By applying ANODE to the positions, proper motions, and photometry of stars observed by Gaia, via machinae obtains a collection of those stars deemed most likely to belong to a stellar stream. We further apply an automated line-finding method based on the Hough transform to search for line-like features in patches of the sky. In this paper, we describe the via machinae algorithm in detail and demonstrate our approach on the prominent stream GD-1. Though some parts of the algorithm are tuned to increase sensitivity to cold streams, the via machinae technique itself does not rely on astrophysical assumptions, such as the potential of the Milky Way or stellar isochrones. This flexibility suggests that it may have further applications in identifying other anomalous structures within the Gaia data set, for example debris flow and globular clusters.

Performance Evaluation of Data-DrivenTechniques for Softwarized and AgnosticManagement of N×N Photonic Switch

The emerging Software Defined Networking (SDN) paradigm paves the way for flexible and automatized management at each layer. The SDN-enabled optical network requires each network element’s software abstraction to enable complete control by the centralized network controller. Nowadays, silicon photonics due to its low energy consumption, low latency, and small footprint is a promising technology for implementing photonic switching topologies, enabling transparent lightpath routing in re-configurable add-drop multiplexers. To this aim, a model for the complete management of photonic switching systems’ control states is fundamental for network control. Typically, photonics-based switches are structured by exploiting the modern technology of Photonic Integrated Circuit (PIC) that enables complex elementary cell structures to be driven individually. Thus PIC switches’ control states are combinations of a large set of elementary controls, and their definition is a challenging task. In this scenario, we propose the use of several data-driven techniques based on Machine Learning (ML) to model the control states of a PIC N×N photonic switch in a completely blind manner. The proposed ML-based techniques are trained and tested in a completely topological and technological agnostic way, and we envision their application in a real-time control plane. The proposed techniques’ scalability and accuracy are validated by considering three different switching topologies: the Honey-Comb Rearrangeable Optical Switch (HCROS), Spanke-Beneš, and the Beneš network. Excellent results in terms of predicting the control states are achieved for all of the considered topologies.

A Flow-based Multi-agent Data Exfiltration Detection Architecture for Ultra-low Latency Networks

Modern network infrastructures host converged applications that demand rapid elasticity of services, increased security, and ultra-fast reaction times. The Tactile Internet promises to facilitate the delivery of these services while enabling new economies of scale for high fidelity of machine-to-machine and human-to-machine interactions. Unavoidably, critical mission systems served by the Tactile Internet manifest high demands not only for high speed and reliable communications but equally, the ability to rapidly identify and mitigate threats and vulnerabilities. This article proposes a novel Multi-Agent Data Exfiltration Detector Architecture (MADEX), inspired by the mechanisms and features present in the human immune system. MADEX seeks to identify data exfiltration activities performed by evasive and stealthy malware that hides malicious traffic from an infected host in low-latency networks. Our approach uses cross-network traffic information collected by agents to effectively identify unknown illicit connections by an operating system subverted. MADEX does not require prior knowledge of the characteristics or behavior of the malicious code or a dedicated access to a knowledge repository. We tested the performance of MADEX in terms of its capacity to handle real-time data and the sensitivity of our algorithm’s classification when exposed to malicious traffic. Experimental evaluation results show that MADEX achieved 99.97% sensitivity, 98.78% accuracy, and an error rate of 1.21% when compared to its best rivals. We created a second version of MADEX, called MADEX level 2, that further improves its overall performance with a slight increase in computational complexity. We argue for the suitability of MADEX level 1 in non-critical environments, while MADEX level 2 can be used to avoid data exfiltration in critical mission systems. To the best of our knowledge, this is the first article in the literature that addresses the detection of rootkits real-time in an agnostic way using an artificial immune system approach while it satisfies strict latency requirements.

A Data-Driven Dimensionality Reduction Approach to Compare and Classify Lipid Force Fields.

Molecular dynamics simulations of all-atom and coarse-grained lipid bilayer models are increasingly used to obtain useful insights for understanding the structural dynamics of these assemblies. In this context, one crucial point concerns the comparison of the performance and accuracy of classical force fields (FFs), which sometimes remains elusive. To date, the assessments performed on different classical potentials are mostly based on the comparison with experimental observables, which typically regard average properties. However, local differences of the structure and dynamics, which are poorly captured by average measurements, can make a difference, but these are nontrivial to catch. Here, we propose an agnostic way to compare different FFs at different resolutions (atomistic, united-atom, and coarse-grained), by means of a high-dimensional similarity metrics built on the framework of Smooth Overlap of Atomic Position (SOAP). We compare and classify a set of 13 FFs, modeling 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) bilayers. Our SOAP kernel-based metrics allows us to compare, discriminate, and correlate different FFs at different model resolutions in an unbiased, high-dimensional way. This also captures differences between FFs in modeling nonaverage events (originating from local transitions), for example, the liquid-to-gel phase transition in dipalmitoylphosphatidylcholine (DPPC) bilayers, for which our metrics allows us to identify nucleation centers for the phase transition, highlighting some intrinsic resolution limitations in implicit versus explicit solvent FFs.

Classifying standard model extensions effectively with precision observables

Effective theories are well established theoretical frameworks to describe the effect of energetically widely separated UV models on observables at lower energy scales. Due to the complexity of the effective theory when taking all the Standard Model symmetries and degrees of freedoms into account, tensioning the entire system in a completely agnostic way against experimental measurements results in constraints on the Wilson Coefficients of the effective operators that either bears little information or challenge intrinsic assumptions imposed on the effective field theory framework. In general, a specific high-scale extension of the Standard Model only induces a subset of all possible operators. Thus, by investigating which operators are induced by different classes of the Standard Model extensions and comparing to which precision observables they contribute, we show that it is possible to obtain an improved understanding of which UV model is realised in nature. We consider 15 UV models which are single scalar field extensions of the Standard Model and compute their dimension-6 operators after integrating out the heavy scalars up to 1-loop level. Only very few of these scenarios remain indistinguishable, while most of the models can be phenomenologically separated from one another. Most of these scenarios possess their own characteristic operator signature. Following the approach outlined here, a comparative analysis of a wide range of models will allow to assess at what level the effective field theory series can be truncated and which experimental measurements to prioritise.