Genomic signal processing (GSP) refers to the use of digital signal processing (DSP) tools for analyzing genomic data such as DNA sequences. A possible application of GSP that has not been fully explored is the computation of the distance between a pair of sequences. In this work we present GAFD, a novel GSP alignment-free distance computation method. We introduce a DNA sequence-to-signal mapping function based on the employment of doublet values, which increases the number of possible amplitude values for the generated signal. Additionally, we explore the use of three DSP distance metrics as descriptors for categorizing DNA signal fragments. Our results indicate the feasibility of employing GAFD for computing sequence distances and the use of descriptors for characterizing DNA fragments.

Mexico, as the center of origin of avocado (Persea americama Mill.), harbors a wide genetic diversity of this species, whose identification may provide the grounds to not only understand its unique population structure and domestication history, but also inform the efforts aimed at its conservation. Although molecular characterization of cultivated avocado germplasm has been studied by several research groups, this had not been the case in Mexico. In order to elucidate the genetic structure of avocado in Mexico and the sustainable use of its genetic resources, 318 avocado accessions conserved in the germplasm collection in the National Avocado Genebank were analyzed using 28 markers [9 expressed sequence tag-Simple Sequence Repeats (SSRs) and 19 genomic SSRs]. Deviation from Hardy Weinberg Equilibrium and high inter-locus linkage disequilibrium were observed especially in drymifolia, and guatemalensis. Total averages of the observed and expected heterozygosity were 0.59 and 0.75, respectively. Although clear genetic differentiation was not observed among 3 botanical races: americana, drymifolia, and guatemalensis, the analyzed Mexican population can be classified into two groups that correspond to two different ecological regions. We developed a core-collection by K-means clustering method. The selected 36 individuals as core-collection successfully represented more than 80% of total alleles and showed heterozygosity values equal to or higher than those of the original collection, despite its constituting slightly more than 10% of the latter. Accessions selected as members of the core collection have now become candidates to be introduced in cryopreservation implying a minimum loss of genetic diversity and a back-up for existing field collections of such important genetic resources.

Genomic signal processing (GSP) is based on the use of digital signal processing methods for the analysis of genomic data. Convolutional neural networks (CNN) are the state-of-the-art machine learning classifiers that have been widely applied to solve complex problems successfully. In this paper, we present a deep learning architecture and a method for the classification of three different functional genome types: coding regions (CDS), long noncoding regions (LNC), and pseudogenes (PSD) in genomic data, based on the use of GSP methods to convert the nucleotide sequence into a graphical representation of the information contained in it. The obtained accuracy scores of 83% and 84% when classifying between CDS vs. LNC and CDS vs. PSD, respectively, indicate the feasibility of employing this methodology for the classification of these types of sequences. The model was not able to differentiate from PSD and LNC. Our results indicate the feasibility of employing CNN with GSP for the classification of these types of DNA data.

Core collections are important tools in genetic resources research and administration. At present, most core collection selection criteria are based on one of the following item characteristics: passport data, genetic markers, or morphological traits, which may lead to inadequate representations of variability in the complete collection. The development of a comprehensive methodology that includes as much element data as possible has been explored poorly. Using a collection of (Setaria italica sbsp. italica (L.) P. Beauv.) as a model, we developed a method for core collection construction based on genotype data and numerical representations of agromorphological traits, thereby improving the selection process. Principal component analysis allows the selection of the most informative discriminators among the various elements evaluated, regardless of whether they are genetic or morphological, thereby providing an adequate criterion for further K-mean clustering. Overall, the core collections of S. italica constructed using only genotype data demonstrated overall better validation scores than other core collections that we generated. However, core collection based on both genotype and agromorphological characteristics represented the overall diversity adequately. The inclusion of both genotype and agromorphological characteristics as a comprehensive dataset in this methodology ensures that agricultural traits are considered in the core collection construction. This approach will be beneficial for genetic resources management and research activities for S. italica as well as other genetic resources.

Alignment-free k-mer-based algorithms in whole genome sequence comparisons remain an ongoing challenge. Here, we explore the possibility to use Topic Modeling for organism whole-genome comparisons. We analyzed 30 complete genomes from three bacterial families by topic modeling. For this, each genome was considered as a document and 13-mer nucleotide representations as words. Latent Dirichlet allocation was used as the probabilistic modeling of the corpus. We where able to identify the topic distribution among analyzed genomes, which is highly consistent with traditional hierarchical classification. It is possible that topic modeling may be applied to establish relationships between genome’s composition and biological phenomena.