Research in Computational Molecular Biology Focused on Comparative Genomics: Selected Papers from RECOMB-CG 2022

With the emergence of single-cell RNA sequencing (scRNA-seq) technology, scientists are able to examine gene expression at single-cell resolution. Analysis of scRNA-seq data has its own challenges, which stem from its high dimensionality. The method of machine learning comes with the potential of gene (feature) selection from the high-dimensional scRNA-seq data. Even though there exist multiple machine learning methods that appear to be suitable for feature selection, such as penalized regression, there is no rigorous comparison of their performances across data sets, where each poses its own challenges. Therefore, in this paper, we analyzed and compared multiple penalized regression methods for scRNA-seq data. Given the scRNA-seq data sets we analyzed, the results show that sparse group lasso (SGL) outperforms the other six methods (ridge, lasso, elastic net, drop lasso, group lasso, and big lasso) using the metrics area under the receiver operating curve (AUC) and computation time. Building on these findings, we proposed a new algorithm for feature selection using penalized regression methods. The proposed algorithm works by selecting a small subset of genes and applying SGL to select the differentially expressed genes in scRNA-seq data. By using hierarchical clustering to group genes, the proposed method bypasses the need for domain-specific knowledge for gene grouping information. In addition, the proposed algorithm provided consistently better AUC for the data sets used. Full article

Recent research in the field of bioinformatics and molecular biology has revealed the immense complexity and uniqueness of microbiomes, while also showcasing the impact of the symbiosis between a microbiome and its host or environment. A core property influencing this process is horizontal gene transfer between members of the bacterial community used to maintain genetic variation. The essential effect of this mechanism is the exposure of genetic information to a wide array of members of the community, creating an additional “layer” of information in the microbiome named the “plasmidome”. From an engineering perspective, introduction of genetic information to an environment must be facilitated into chosen species which will be able to carry out the desired effect instead of competing and inhibiting it. Moreover, this process of information transfer imposes concerns for the biosafety of genetic engineering of microbiomes as exposure of genetic information into unwanted hosts can have unprecedented ecological impacts. Current technologies are usually experimentally developed for a specific host/environment, and only deal with the transformation process itself at best, ignoring the impact of horizontal gene transfer and gene-microbiome interactions that occur over larger periods of time in uncontrolled environments. The goal of this research was to design new microbiome-specific versions of engineered genetic information, providing an additional layer of compatibility to existing engineering techniques. The engineering framework is entirely computational and is agnostic to the selected microbiome or gene by reducing the problem into the following set up: microbiome species can be defined as wanted or unwanted hosts of the modification. Then, every element related to gene expression (e.g., promoters, coding regions, etc.) and regulation is individually examined and engineered by novel algorithms to provide the defined expression preferences. Additionally, the synergistic effect of the combination of engineered gene blocks facilitates robustness to random mutations that might occur over time. This method has been validated using both computational and experimental tools, stemming from the research done in the iGEM 2021 competition, by the TAU group. Full article

Phylogenetic placement, used widely in ecological analyses, seeks to add a new species to an existing tree. A deep learning approach was previously proposed to estimate the distance between query and backbone species by building a map from gene sequences to a high-dimensional space that preserves species tree distances. They then use a distance-based placement method to place the queries on that species tree. In this paper, we examine the appropriate geometry for faithfully representing tree distances while embedding gene sequences. Theory predicts that hyperbolic spaces should provide a drastic reduction in distance distortion compared to the conventional Euclidean space. Nevertheless, hyperbolic embedding imposes its own unique challenges related to arithmetic operations, exponentially-growing functions, and limited bit precision, and we address these challenges. Our results confirm that hyperbolic embeddings have substantially lower distance errors than Euclidean space. However, these better-estimated distances do not always lead to better phylogenetic placement. We then show that the deep learning framework can be used not just to place on a backbone tree but to update it to obtain a fully resolved tree. With our hyperbolic embedding framework, species trees can be updated remarkably accurately with only a handful of genes. Full article

Phylogenetic identification of unknown sequences by placing them on a tree is routinely attempted in modern ecological studies. Such placements are often obtained from incomplete and noisy data, making it essential to augment the results with some notion of uncertainty. While the standard likelihood-based methods designed for placement naturally provide such measures of uncertainty, the newer and more scalable distance-based methods lack this crucial feature. Here, we adopt several parametric and nonparametric sampling methods for measuring the support of phylogenetic placements that have been obtained with the use of distances. Comparing the alternative strategies, we conclude that nonparametric bootstrapping is more accurate than the alternatives. We go on to show how bootstrapping can be performed efficiently using a linear algebraic formulation that makes it up to 30 times faster and implement this optimized version as part of the distance-based placement software APPLES. By examining a wide range of applications, we show that the relative accuracy of maximum likelihood (ML) support values as compared to distance-based methods depends on the application and the dataset. ML is advantageous for fragmentary queries, while distance-based support values are more accurate for full-length and multi-gene datasets. With the quantification of uncertainty, our work fills a crucial gap that prevents the broader adoption of distance-based placement tools. Full article