Back

De novo clustering of long-read amplicons improves phylogenetic insight into microbiome data

Hui, Y.; Nielsen, D. S.; Krych, L.

2023-11-26 bioinformatics
10.1101/2023.11.26.568539 bioRxiv
Show abstract

Long-read amplicon profiling through read classification limits phylogenetic analysis of amplicons while community analysis of multicopy genes, relying on unique molecular identifier (UMI) corrections, often demands deep sequencing. To address this, we present a long amplicon consensus analysis (LACA) workflow employing multiple de novo clustering approaches based on sequence dissimilarity. LACA controls the average error rate of corrected sequences below 1% for the Oxford Nanopore Technologies (ONT) R9.4.1 and ONT R10.3 data, 0.2% for ONT R10.4.1, and 0.1% for high-accuracy ONT Duplex and Pacific Biosciences (PacBio) circular consensus sequencing (CCS) data in both simulated 16S rRNA and real 16-23S rRNA amplicon datasets. In high-accuracy PacBio CCS data, the clustering-based correction matched UMI correction, while outperforming 4xUMI correction in noisy ONT R10.3 and R9.4.1 data. Notably, LACA preserved phylogenetic fidelity in long operational taxonomic units and enhanced microbiome-wide phenotype characterization for synthetic mock communities and human vaginal samples.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

1
Nature Communications
○ 5641 papers in training set
Top 5%
26.0%
Similar papers in this journal
2
Microbiome
○ 154 papers in training set
Top 0.4%
7.7%
Similar papers in this journal
3
Genome Biology
○ 637 papers in training set
Top 2%
7.1%
Similar papers in this journal
4
Scientific Reports
○ 3612 papers in training set
Top 17%
5.4%
5
mSystems
● 394 papers in training set
Top 2%
5.4%
Similar papers in this journal
50% of probability mass above
6
Communications Biology
○ 993 papers in training set
Top 2%
5.4%
Similar papers in this journal
7
Briefings in Bioinformatics
◐ 354 papers in training set
Top 2%
3.9%
8
Microbiology Spectrum
● 469 papers in training set
Top 4%
3.9%
9
Molecular Ecology Resources
○ 171 papers in training set
Top 0.8%
3.1%
10
Genome Research
● 468 papers in training set
Top 3%
2.3%
11
Nucleic Acids Research
◐ 1281 papers in training set
Top 8%
2.1%
12
GigaScience
◐ 212 papers in training set
Top 2%
1.7%
13
BMC Bioinformatics
○ 457 papers in training set
Top 4%
1.5%
14
iMeta
○ 10 papers in training set
Top 0.1%
1.3%
15
Computational and Structural Biotechnology Journal
● 242 papers in training set
Top 4%
1.3%
16
Frontiers in Microbiology
○ 427 papers in training set
Top 7%
1.1%
17
Microbial Genomics
● 225 papers in training set
Top 2%
1.1%
18
Analytical Chemistry
● 218 papers in training set
Top 2%
1.1%
19
PLOS Computational Biology
● 1863 papers in training set
Top 17%
1.1%
20
BMC Genomics
○ 406 papers in training set
Top 7%
1.0%
21
iScience
○ 1154 papers in training set
Top 33%
1.0%
22
NAR Genomics and Bioinformatics
◐ 242 papers in training set
Top 4%
1.0%
23
PeerJ
◐ 308 papers in training set
Top 9%
1.0%
24
PLOS ONE
● 5266 papers in training set
Top 57%
1.0%
25
Scientific Data
○ 209 papers in training set
Top 3%
0.8%
26
Nature Computational Science
○ 55 papers in training set
Top 2%
0.6%

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.