arXiv research

Detects adversarial samples in DNNs using model mutation testing.

Mutation Validation method assesses model fit using mutated training labels without validation sets.

PANDA predicts protein binding affinity changes from sequences, outperforming existing methods.

Study compares mutation validation and cross-validation for model selection.

MutaGAN predicts mutations of evolving protein populations using GANs.

Modeling correlated mutations in cancer for personalized treatment.

We classify the Montesinos links up to mutation and 5-move equivalence, and obtain from this a Jones and Kauffman polynomial test for a Montesinos link.

This paper presents details of our winning solutions to the task IV of NIPS 2017 Competition Track entitled Classifying Clinically Actionable Genetic Mutations. The machine learning task aims to classify genetic mutations based on text evidence from clinical literature with promising performance. We develop a novel mul…

We apply our statistically deterministic machine learning/clustering algorithm *K-means (recently developed in https://ssrn.com/abstract=2908286) to 10,656 published exome samples for 32 cancer types. A majority of cancer types exhibit mutation clustering structure. Our results are in-sample stable. They are also out-o…

Improved genetic programming by optimizing mutation operators for continuous program search.

Genus 2 mutation is the process of cutting a 3-manifold along an embedded closed genus 2 surface, twisting by the hyper-elliptic involution, and gluing back. This paper compares genus 2 mutation with the better-known Conway mutation in the context of knots in the 3-sphere. Despite the fact that any Conway mutation can …

Researchers propose improved multivariate prediction models for HIV drug resistance.

Paper tackles cancer mutation data challenges by creating useful low-dimensional representations.

EA improves clustering efficiency using crossover and mutation.

We define a link homology theory that is readily seen to be both isomorphic to reduced odd Khovanov homology and fully determined by data impervious to Conway mutation. This gives an elementary proof that odd Khovanov homology is mutation invariant, and therefore that mod 2 Khovanov homology is mutation invariant. We a…

We give a new, elementary proof that Khovanov homology with $\mathbb{Z}/2\mathbb{Z}$--coefficients is invariant under Conway mutation. This proof also gives a strategy to prove Baldwin and Levine's conjecture that $δ$--graded knot Floer homology is mutation--invariant. Using the Clifford module structure on $\widetilde…

Mathematician summarizes protein geometry and mutation effects.

We study the effect of mutation on link concordance and 3-manifolds. We show that the set of links concordant to sublinks of homology boundary links is not closed under positive mutation. We show that mutation does not preserve homology cobordism classes of 3-manifolds. A significant consequence is that there exist 3-m…

Deep generative model for healthcare data identifies coherent substructures and mutational clusters.

Mutation graph of support τ-tilting modules over skew-gentle algebras is connected.

Khovanov homology invariant under Conway mutation.

New invariant measures loop iterations in algebraic structures.

Recently, it has been shown that deep neural networks (DNN) are subject to attacks through adversarial samples. Adversarial samples are often crafted through adversarial perturbation, i.e., manipulating the original sample with minor modifications so that the DNN model labels the sample incorrectly. Given that it is al…

The paper studies alternating links in thickened surfaces using flow lattices and disc mutations.

VEGN uses graph neural networks to predict disease-causing mutations from genetic variants.

Paper tackles gene mutation prediction for HCC using multi-instance multi-label learning.

Deep Learning can significantly benefit cancer proteomics and genomics. In this study, we attempt to determine a set of critical proteins that are associated with the FLT3-ITD mutation in newly-diagnosed acute myeloid leukemia patients. A Deep Learning network consisting of autoencoders forming a hierarchical model fro…

We show that instanton knot homology is mutation-invariant, as a consequence of earlier work of the third author.

Study shows Conway mutation preserves a specific link invariant.

We construct finite volume hyperbolic manifolds with large symmetry groups. The construction makes use of the presentations of finite Coxeter groups provided by Barot and Marsh and involves mutations of quivers and diagrams defined in the theory of cluster algebras. We generalize our construction by assigning to every …

Active learning speeds up antibody affinity prediction.

We prove that generalized mutation preserves several geometric invariants such as the volume and Goncharov invariant of Q-rank 1 locally symmetric spaces.

EDAs with matrix transpose improve Bayesian structure learning performance.

The study explores properties and mutations in oriented matroids, proving new results on Euclidean and non-Euclidean structures.

The study of high-throughput genomic profiles from a pharmacogenomics viewpoint has provided unprecedented insights into the oncogenic features modulating drug response. A recent screening of ~1,000 cancer cell lines to a collection of anti-cancer drugs illuminated the link between genotypes and vulnerability. However,…

We show that the reduced sl(n) homology defined by Khovanov and Rozansky is invariant under component-preserving positive mutation when n is odd.

We prove that Khovanov homology and Lee homology with coefficients in $\mathbb{F}_2$ are invariant under component-preserving link mutations.

In many real life problems, objects are described by large number of binary features. For instance, documents are characterized by presence or absence of certain keywords; cancer patients are characterized by presence or absence of certain mutations etc. In such cases, grouping together similar objects/profiles based o…

We give a short topological proof for Rubermans Theorem about mutation and volume, using the Maskit combination theorem and the homology of the linear group.

SynGAN generates synthetic network attacks to improve NIDS effectiveness.

Deep learning improves tumor type classification accuracy.

Proteins are commonly used by biochemical industry for numerous processes. Refining these proteins' properties via mutations causes stability effects as well. Accurate computational method to predict how mutations affect protein stability are necessary to facilitate efficient protein design. However, accuracy of predic…

In this paper, we explicitly construct large classes of incommensurable hyperbolic knot complements with the same volume and the same initial (complex) length spectrum. Furthermore, we show that these knot complements are the only knot complements in their respective commensurabiltiy classes by analyzing their cusp sha…

For the alternating knots or links, mutations do not change the arc index. In the case of nonalternating knots, some semi-alternating knots or links have this property. We mainly focus on the problem of mutation invariance of the arc index for nonalternating knots which are not semi-alternating. In this paper, we found…

We describe and investigate a connection between the topology of isolated singularities of plane curves and the mutation equivalence, in the sense of cluster algebra theory, of the quivers associated with their morsifications.

An evolutionary algorithm separates mixed DNA profiles in forensic genetics.

Given a hyperbolic knot, we prove that the Reidemeister torsion of any lift of the holonomy to SL(2,C) is invariant under mutation along a Conway sphere.

Mutation is an operation on 3-manifolds containing an embedded surface of genus 2. It is defined by cutting along the surface and regluing using the `hyperelliptic' involution, and is known to preserve many 3-manifold invariants. I show that mutation of a homology 3-sphere preserves its (instanton) Floer homology, and …