PubMed Health⌕ Search

Biomedical subjects

Degui Zhi

Publications and source records attributed to Degui Zhi.

8 recordsLinked to original sources

RLBWT-based LCP computation in compressed space for terabase-scale pangenome analysis.

MOTIVATION: Lossless full text indexes are utilized in a myriad of applications in bioinformatics. The continuously decreasing cost of generating biological data has resulted in the need to build full text indexes on biological datasets of increasing size. Many compressed full text indexes have been developed to address this problem. In particular, run-length Burrows-Wheeler transform (RLBWT) based compressed full text indexes have seen wide development and adoption. However, the construction of these RLBWT-based compressed full text indexes is still computationally expensive, sometimes prohibitively so, even for current dataset sizes. RESULTS: Therefore, we present algorithms for the construction of RLBWT-based compressed full text indexes and their supporting data structures in compressed space. The algorithms have a space complexity of O(r) words and run in O(n) time for repetitive datasets, where r is the number of runs in the BWT, n is the length of the text, and repetitive datasets implies nr∈Ω(log n). We provide the first algorithm to compute LCP-related information for repetitive datasets in optimal time and O(r) space, greatly reducing memory requirements. The key idea behind this algorithm is the utilization of r samples of the inverse suffix array at regular intervals. For example, on the Human Pangenome Reference Consortium Release 2 dataset, this reduces peak memory from 2135 GiB to 170 GiB (12.6x reduction) compared to the previous best method (pfp-thresholds). AVAILABILITY AND IMPLEMENTATION: The implementation is available at https://github.com/ucfcbb/TeraTools.

Algorithms↗

GWAS for Periodontitis Phenotypes Using Multi-Ancestry All of Us Research Platform.

Periodontitis is a multifactorial inflammatory disease whose pathogenesis is associated with intricate interactions between genetic and environmental factors. Leveraging electronic health records data from the All of Us Research Program, we stratified periodontitis by clinically relevant dimensions: stage, grade, and extent. Based on these phenotypes, we performed a multi-ancestry genome-wide association study, focusing on predominant ancestry populations of African, European, and Admixed American. Our study cohort comprised 3,881 periodontitis patients and a control group of 10,760 patients with dental caries and without periodontitis. Ancestry-specific GWAS revealed significant genetic associations (P<5&#xd7;10-8) in periodontitis grade phenotypes at the LINC00294 and CLMN loci in the African ancestry population and also confirmed via the multi-ancestry meta-analysis. In addition, the XYLT1 locus emerged as a significant signal associated with periodontitis grade phenotype in the admixed American GWAS. Our GWAS comparing periodontitis to dental caries in the admixed American population identified several significant loci, including RABGAP1L, previously linked to immune regulation, DCHS2, a cadherin-related gene involved in bone mineralization and tissue morphogenesis, and OSTM1, known to be crucial for bone remodeling. The findings of our study highlight the potential of integrating EHR and genomic data from large-scale biobanks to achieve informative dental phenotyping, uncover novel molecular insights into periodontal disease, and personalize treatment approaches.

Journal Article↗

Representing and comparing protein structures as paths in three-dimensional space.

BACKGROUND: Most existing formulations of protein structure comparison are based on detailed atomic level descriptions of protein structures and bypass potential insights that arise from a higher-level abstraction. RESULTS: We propose a structure comparison approach based on a simplified representation of proteins that describes its three-dimensional path by local curvature along the generalized backbone of the polypeptide. We have implemented a dynamic programming procedure that aligns curvatures of proteins by optimizing a defined sum turning angle deviation measure. CONCLUSION: Although our procedure does not directly optimize global structural similarity as measured by RMSD, our benchmarking results indicate that it can surprisingly well recover the structural similarity defined by structure classification databases and traditional structure alignment programs. In addition, our program can recognize similarities between structures with extensive conformation changes that are beyond the ability of traditional structure alignment programs. We demonstrate the applications of procedure to several contexts of structure comparison. An implementation of our procedure, CURVE, is available as a public webserver.

Algorithms↗

Sequence correlation between neighboring Alu instances suggests post-retrotransposition sequence exchange due to Alu gene conversion.

Alu elements constitute 10% of the human genome. Alu mobilization is important in the evolution of the human genome. While retrotransposition is the primary pathway of Alu mobilization, Alu gene conversion has been postulated as a secondary pathway for Alu mobilization in human genome. However, the mode and tempo of Alu gene conversions remain a mystery due to lack of sensitive statistical methods. In this paper, we present the first study on sequence correlation between Alu instances, measured by the number of shared mutations away from the Alu consensus, or co-mutations. Our analysis reveals a significantly elevated co-mutation rate between Alu instances that are located in close proximity along a chromosome. This effect is more pronounced outside Alu subfamily diagnostic positions. This effect peaks among immediately adjacent Alu instances, diminishes quickly in increasing distances between Alu instances, and vanishes beyond 5000 bp. Our results suggest that this effect reflects post-retrotransposition sequence exchanges between Alu instances, mainly due to Alu gene conversions.

Alu Elements↗

AliWABA: alignment on the web through an A-Bruijn approach.

UNLABELLED: Multiple sequence alignment programs are an invaluable tool in computational biology. A-Bruijn Alignment (ABA) is a method for multiple sequence alignment that represents an alignment as a directed graph and has proved useful in aligning nucleotide and amino acid sequences that are composed of repeated and shuffled subsequences. AliWABA is a web server that provides tools to generate alignments with ABA, visualize the resulting ABA graphs and extract subsequences from ABA graphs. AliWABA greatly simplifies the problem of analyzing multiple sequences for local similarities that may be reordered, as is common with the domain architectures of proteins. To facilitate the analysis of protein domains, AliWABA provides direct querying of the Conserved Domain Database. AVAILABILITY: http://aba.nbcr.net/

Algorithms↗

Identifying repeat domains in large genomes.

We present a graph-based method for the analysis of repeat families in a repeat library. We build a repeat domain graph that decomposes a repeat library into repeat domains, short subsequences shared by multiple repeat families, and reveals the mosaic structure of repeat families. Our method recovers documented mosaic repeat structures and suggests additional putative ones. Our method is useful for elucidating the evolutionary history of repeats and annotating de novo generated repeat libraries.

Algorithms↗

Identification of highly specific localized sequence motifs in human ribosomal protein gene promoters.

For ribosomal protein (RP) genes the start of transcription is rigidly controlled to maintain the 5'-TOP signal on the messenger RNA. The responsible regulatory mechanism is not yet fully understood. Careful comparative analysis of their proximal promoter sequences reveals common characteristics and thus provides clues to the underlying mechanism. We have extracted the proximal promoters of the 80 human cytosolic ribosomal protein genes together with the orthologous mouse sequences. After annotating the set with transcription factor binding sites based on the available literature, we searched for over-represented sequence motifs. We uncovered a novel motif that is localized at a fixed distance downstream to the transcription start. 31 out of the 80 promoters contain the motif in the same orientation around position +62 (standard deviation 6). A second evolutionary conserved and palindromic motif is found 13 times in the RP promoter set, 9 instances of which are located upstream around position -40. In addition, we see a characteristic profile of the GC-content and of the CpG dinucleotide frequencies. Our results support a model for the transcription of ribosomal protein genes in which the maintenance of the accurate start of transcription is provided by specific transcription factors. Such a factor binds the target DNA at a fixed location relative to the TSS, and possibly interacts directly with the basal transcription machinery.

Amino Acid Motifs↗

A novel method for multiple alignment of sequences with repeated and shuffled elements.

We describe ABA (A-Bruijn alignment), a new method for multiple alignment of biological sequences. The major difference between ABA and existing multiple alignment methods is that ABA represents an alignment as a directed graph, possibly containing cycles. This representation provides more flexibility than does a traditional alignment matrix or the recently introduced partial order alignment (POA) graph by allowing a larger class of evolutionary relationships between the aligned sequences. Our graph representation is particularly well-suited to the alignment of protein sequences with shuffled and/or repeated domain structure, and allows one to construct multiple alignments of proteins containing (1) domains that are not present in all proteins, (2) domains that are present in different orders in different proteins, and (3) domains that are present in multiple copies in some proteins. In addition, ABA is useful in the alignment of genomic sequences that contain duplications and inversions. We provide several examples illustrating the applications of ABA.

Algorithms↗