More Like this

1. Privacy and anonymity for multilayer networks: A reflection

   Abhishel Santraa, Kiran Mukunda ( July 2023 , IEEEBigDataService)

   AbstractÐPrivacy of data as well as providing anonymization of data for various kinds of analysis have been addressed in the context of tabular transactional data which was mainstream. With the advent of the Internet and social networks, there is an emphasis on using different kinds of graphs for modeling and analysis. In addition to single graphs, the use of MultiLayer Networks (or MLNs) for modeling and analysis is becoming popular for complex data having multiple types of entities and relationships. They provide a better understanding of data as well as flexibility and efficiency of analysis. In this article, we understand the provenance of data privacy and some of the thinking on extending it to graph data models. We will focus on the issues of data privacy for models that are different from traditional data models and discuss alternatives. We will also consider privacy from a visualization perspective as we have developed a community Dashboard for MLN generation, analysis, and visualization based on our research. 

   more » « less
2. phenotools: An r package for visualizing and analysing phenomic datasets

   https://doi.org/10.1111/2041-210X.13217  

   Eliason, Chad M. ; Edwards, Scott V. ; Clarke, Julia A. ( July 2019 , Methods in Ecology and Evolution)

   Phenotypic data are crucial for understanding genotype–phenotype relationships, assessing the tree of life and revealing trends in trait diversity over time. Large‐scale description of whole organisms for quantitative analyses (phenomics) presents several challenges, and technological advances in the collection of genomic data outpace those for phenomic data. Reasons for this disparity include the time‐consuming and expensive nature of collecting discrete phenotypic data and mining previously published data on a given species (both often requiring anatomical expertise across taxa), and computational challenges involved with analysing high‐dimensional datasets.

   One approach to building approximations of organismal phenomes is to combine published datasets of discrete characters assembled for phylogenetic analyses into a phenomic dataset. Despite a wealth of legacy datasets in the literature for many groups, relatively few methods exist for automating the assembly, analysis, and visualization of phenomic datasets in phylogenetic contexts. Here, we introduce a newrpackagephenotoolsfor integrating (fusing original or legacy datasets), curating (finding and removing duplicates) and visualizing phenomic datasets.

   We demonstrate the utility of the proposed toolkit with a morphological dataset for flightless birds and two morphological datasets for theropod dinosaurs and provide recommendations for character construction to maximize accessibility in future workflows. Visualization tools allow rapid identification of anatomical subregions with difficult or problematic histories of homology.

   We anticipate these tools aiding automation of the assembly and visualization of phenomic datasets to inform evolutionary relationships and rates of phenotypic evolution.

    

   more » « less
3. Thunor: visualization and analysis of high-throughput dose–response datasets

   https://doi.org/10.1093/nar/gkab424  

   Lubbock, Alexander L ; Harris, Leonard A ; Quaranta, Vito ; Tyson, Darren R ; Lopez, Carlos F ( May 2021 , Nucleic Acids Research)

   Abstract High-throughput cell proliferation assays to quantify drug-response are becoming increasingly common and powerful with the emergence of improved automation and multi-time point analysis methods. However, pipelines for analysis of these datasets that provide reproducible, efficient, and interactive visualization and interpretation are sorely lacking. To address this need, we introduce Thunor, an open-source software platform to manage, analyze, and visualize large, dose-dependent cell proliferation datasets. Thunor supports both end-point and time-based proliferation assays as input. It provides a simple, user-friendly interface with interactive plots and publication-quality images of cell proliferation time courses, dose–response curves, and derived dose–response metrics, e.g. IC50, including across datasets or grouped by tags. Tags are categorical labels for cell lines and drugs, used for aggregation, visualization and statistical analysis, e.g. cell line mutation or drug class/target pathway. A graphical plate map tool is included to facilitate plate annotation with cell lines, drugs and concentrations upon data upload. Datasets can be shared with other users via point-and-click access control. We demonstrate the utility of Thunor to examine and gain insight from two large drug response datasets: a large, publicly available cell viability database and an in-house, high-throughput proliferation rate dataset. Thunor is available from www.thunor.net. 

   more » « less
4. Ferret: Reviewing Tabular Datasets for Manipulation

   https://doi.org/10.1111/cgf.14822  

   Lange, Devin ; Sahai, Shaurya ; Phillips, Jeff M. ; Lex, Alexander ( June 2023 , Computer Graphics Forum)

   How do we ensure the veracity of science? The act of manipulating or fabricating scientific data has led to many high‐profile fraud cases and retractions. Detecting manipulated data, however, is a challenging and time‐consuming endeavor. Automated detection methods are limited due to the diversity of data types and manipulation techniques. Furthermore, patterns automatically flagged as suspicious can have reasonable explanations. Instead, we propose a nuanced approach where experts analyze tabular datasets, e.g., as part of the peer‐review process, using a guided, interactive visualization approach. In this paper, we present an analysis of how manipulated datasets are created and the artifacts these techniques generate. Based on these findings, we propose a suite of visualization methods to surface potential irregularities. We have implemented these methods in Ferret, a visualization tool for data forensics work. Ferret makes potential data issues salient and provides guidance on spotting signs of tampering and differentiating them from truthful data.

    

   more » « less
5. Issues in the Reproducibility of Deep Learning Results

   https://doi.org/10.1109/SPMB47826.2019.9037840  

   Jean-Paul, S. ; Elseify, T. ; Obeid, I. ; Picone, J. ( December 2019 , IEEE Signal Processing in Medicine and Biology Symposium (SPMB))

   Obeid, I. ; Selesnik, I. ; Picone, J. (Ed.)

   The Neuronix high-performance computing cluster allows us to conduct extensive machine learning experiments on big data [1]. This heterogeneous cluster uses innovative scheduling technology, Slurm [2], that manages a network of CPUs and graphics processing units (GPUs). The GPU farm consists of a variety of processors ranging from low-end consumer grade devices such as the Nvidia GTX 970 to higher-end devices such as the GeForce RTX 2080. These GPUs are essential to our research since they allow extremely compute-intensive deep learning tasks to be executed on massive data resources such as the TUH EEG Corpus [2]. We use TensorFlow [3] as the core machine learning library for our deep learning systems, and routinely employ multiple GPUs to accelerate the training process. Reproducible results are essential to machine learning research. Reproducibility in this context means the ability to replicate an existing experiment – performance metrics such as error rates should be identical and floating-point calculations should match closely. Three examples of ways we typically expect an experiment to be replicable are: (1) The same job run on the same processor should produce the same results each time it is run. (2) A job run on a CPU and GPU should produce identical results. (3) A job should produce comparable results if the data is presented in a different order. System optimization requires an ability to directly compare error rates for algorithms evaluated under comparable operating conditions. However, it is a difficult task to exactly reproduce the results for large, complex deep learning systems that often require more than a trillion calculations per experiment [5]. This is a fairly well-known issue and one we will explore in this poster. Researchers must be able to replicate results on a specific data set to establish the integrity of an implementation. They can then use that implementation as a baseline for comparison purposes. A lack of reproducibility makes it very difficult to debug algorithms and validate changes to the system. Equally important, since many results in deep learning research are dependent on the order in which the system is exposed to the data, the specific processors used, and even the order in which those processors are accessed, it becomes a challenging problem to compare two algorithms since each system must be individually optimized for a specific data set or processor. This is extremely time-consuming for algorithm research in which a single run often taxes a computing environment to its limits. Well-known techniques such as cross-validation [5,6] can be used to mitigate these effects, but this is also computationally expensive. These issues are further compounded by the fact that most deep learning algorithms are susceptible to the way computational noise propagates through the system. GPUs are particularly notorious for this because, in a clustered environment, it becomes more difficult to control which processors are used at various points in time. Another equally frustrating issue is that upgrades to the deep learning package, such as the transition from TensorFlow v1.9 to v1.13, can also result in large fluctuations in error rates when re-running the same experiment. Since TensorFlow is constantly updating functions to support GPU use, maintaining an historical archive of experimental results that can be used to calibrate algorithm research is quite a challenge. This makes it very difficult to optimize the system or select the best configurations. The overall impact of all of these issues described above is significant as error rates can fluctuate by as much as 25% due to these types of computational issues. Cross-validation is one technique used to mitigate this, but that is expensive since you need to do multiple runs over the data, which further taxes a computing infrastructure already running at max capacity. GPUs are preferred when training a large network since these systems train at least two orders of magnitude faster than CPUs [7]. Large-scale experiments are simply not feasible without using GPUs. However, there is a tradeoff to gain this performance. Since all our GPUs use the NVIDIA CUDA® Deep Neural Network library (cuDNN) [8], a GPU-accelerated library of primitives for deep neural networks, it adds an element of randomness into the experiment. When a GPU is used to train a network in TensorFlow, it automatically searches for a cuDNN implementation. NVIDIA’s cuDNN implementation provides algorithms that increase the performance and help the model train quicker, but they are non-deterministic algorithms [9,10]. Since our networks have many complex layers, there is no easy way to avoid this randomness. Instead of comparing each epoch, we compare the average performance of the experiment because it gives us a hint of how our model is performing per experiment, and if the changes we make are efficient. In this poster, we will discuss a variety of issues related to reproducibility and introduce ways we mitigate these effects. For example, TensorFlow uses a random number generator (RNG) which is not seeded by default. TensorFlow determines the initialization point and how certain functions execute using the RNG. The solution for this is seeding all the necessary components before training the model. This forces TensorFlow to use the same initialization point and sets how certain layers work (e.g., dropout layers). However, seeding all the RNGs will not guarantee a controlled experiment. Other variables can affect the outcome of the experiment such as training using GPUs, allowing multi-threading on CPUs, using certain layers, etc. To mitigate our problems with reproducibility, we first make sure that the data is processed in the same order during training. Therefore, we save the data from the last experiment and to make sure the newer experiment follows the same order. If we allow the data to be shuffled, it can affect the performance due to how the model was exposed to the data. We also specify the float data type to be 32-bit since Python defaults to 64-bit. We try to avoid using 64-bit precision because the numbers produced by a GPU can vary significantly depending on the GPU architecture [11-13]. Controlling precision somewhat reduces differences due to computational noise even though technically it increases the amount of computational noise. We are currently developing more advanced techniques for preserving the efficiency of our training process while also maintaining the ability to reproduce models. In our poster presentation we will demonstrate these issues using some novel visualization tools, present several examples of the extent to which these issues influence research results on electroencephalography (EEG) and digital pathology experiments and introduce new ways to manage such computational issues. 

   more » « less