arXiv subjects

Zanoni Dias

Publications and source records attributed to Zanoni Dias.

6 recordsLinked to original sources

arXiv2025

Approximations for the Weighted Reversal, Transposition, and Indel Distance Problem with Intergenic Region Information

Genome rearrangement distances are an established method in genome comparison. Works in this area may include various rearrangement operations representing large-scale mutations, gene orientation information, the number of nucleotides in intergenic regions, and weights reflecting the expected frequency of each operation. In this article, we model genomes containing at most one copy of each gene by considering gene sequences, with orientations, and representing intergenic regions according to their nucleotide lengths. We looked at a problem called Weighted Reversal, Transposition, and Indel Distance, which seeks the minimal cost sequence composed by the rearrangement operations of reversals, transposition, and indels, capable of transforming one genome into another. We leverage a structure called Labeled Intergenic Breakpoint Graph to show an algorithm for that problem with guaranteed approximations considering some sets of weights for the operations.

cs.DS↗

arXiv2025

Scaling Up ESM2 Architectures for Long Protein Sequences Analysis: Long and Quantized Approaches

Various approaches utilizing Transformer architectures have achieved state-of-the-art results in Natural Language Processing (NLP). Based on this success, numerous architectures have been proposed for other types of data, such as in biology, particularly for protein sequences. Notably among these are the ESM2 architectures, pre-trained on billions of proteins, which form the basis of various state-of-the-art approaches in the field. However, the ESM2 architectures have a limitation regarding input size, restricting it to 1,022 amino acids, which necessitates the use of preprocessing techniques to handle sequences longer than this limit. In this paper, we present the long and quantized versions of the ESM2 architectures, doubling the input size limit to 2,048 amino acids.

cs.LG↗

arXiv2025

Heuristics based on Adjacency Graph Packing for DCJ Distance Considering Intergenic Regions

In this work, we explore heuristics for the Adjacency Graph Packing problem, which can be applied to the Double Cut and Join (DCJ) Distance Problem. The DCJ is a rearrangement operation and the distance problem considering it is a well established method for genome comparison. Our heuristics will use the structure called adjacency graph adapted to include information about intergenic regions, multiple copies of genes in the genomes, and multiple circular or linear chromosomes. The only required property from the genomes is that it must be possible to turn one into the other with DCJ operations. We propose one greedy heuristic and one heuristic based on Genetic Algorithms. Our experimental tests in artificial genomes show that the use of heuristics is capable of finding good results that are superior to a simpler random strategy.

q-bio.QM↗

arXiv2024

P-NOC: adversarial training of CAM generating networks for robust weakly supervised semantic segmentation priors

Weakly Supervised Semantic Segmentation (WSSS) techniques explore individual regularization strategies to refine Class Activation Maps (CAMs). In this work, we first analyze complementary WSSS techniques in the literature, their segmentation properties, and the conditions in which they are most effective. Based on these findings, we devise two new techniques: P-NOC and CCAM-H. In the first, we promote the conjoint training of two adversarial CAM generating networks: the generator, which progressively learns to erase regions containing class-specific features, and a discriminator, which is refined to gradually shift its attention to new class discriminant features. In the latter, we employ the high quality pseudo-segmentation priors produced by P-NOC to guide the learning to saliency information in a weakly supervised fashion. Finally, we employ both pseudo-segmentation priors and pseudo-saliency proposals in the random walk procedure, resulting in higher quality pseudo-semantic segmentation masks, and competitive results with the state of the art.

cs.CV↗

arXiv2022

Algorithms for the Maximum Eulerian Cycle Decomposition Problem

Given an Eulerian graph G, in the Maximum Eulerian Cycle Decomposition problem, we are interested in finding a collection of edge-disjoint cycles {E_1, E_2, ..., E_k} in G such that all edges of G are in exactly one cycle and k is maximum. We present an algorithm to solve the pricing problem of a column generation Integer Linear Programming (ILP) model introduced by Lancia and Serafini (2016). Furthermore, we propose a greedy heuristic, which searches for minimum size cycles starting from a random vertex, and a heuristic based on partially solving the ILP model. We performed tests comparing the three approaches in relation to the quality of solutions and execution time, using distinct sets of Eulerian graphs, each set grouping graphs with different numbers of vertices and edges. Our experimental results show that the ILP based heuristic outperforms the other methods.

cs.DS↗

arXiv2022

On the Complexity of Some Variations of Sorting by Transpositions

One of the main challenges in Computational Biology is to find the evolutionary distance between two organisms. In the field of comparative genomics, one way to estimate such distance is to find a minimum cost sequence of rearrangements (large scale mutations) needed to transform one genome into another, which is called the rearrangement distance. In the past decades, these problems were studied considering many types of rearrangements (such as reversals, transpositions, transreversals, and revrevs) and considering the same weight for all rearrangements, or different weights depending on the types of rearrangements. The complexity of the problems involving reversals, transpositions, and both rearrangements is known, even though the hardness proof for the problem combining reversals and transpositions was recently given. In this paper, we enhance the knowledge for these problems by proving that models involving transpositions alongside reversals, transreversals, and revrevs are NP-hard, considering weights w1 for reversals and w2 for the other rearrangements such that w2/w1 is less than or equal to 1.5. In addition, we address a cost function related to the number of fragmentations caused by a rearrangement, proving that the problem of finding a minimum cost sorting sequence, considering the fragmentation cost function with some restrictions, is NP-hard for transpositions and the combination of reversals and transpositions.

cs.CC↗