Quorum sensing switches on plasmid transfer, as it does in Agrobacterium. Each loop, the genes land in a different order. Schematic, not to scale.

Bacterial genomes & other rabbit holes

Kinley Tenzin

Ph.D. Student Biology Kansas State University Manhattan, Kansas

I'm interested in the combinatorics of mobile genetic elements, how they move, mix, and reshape bacterial genomes. I'm still figuring out how to study all of this properly. Fortunately, I have Tom Platt helping me along the way. My consulting days left me obsessed with finding signal in noise. These days, I mostly nerd out over false discovery rates (FDR) and how easily p-values can fool us.

Outside of research, I still play Counter-Strike 1.6.

Ongoing Research Publications Contact

tenzin12@ksu.edu ORCID 0009-0009-2417-6880 GitHub

Fig. 1 · Kalman filter A little signal in the noise

Noisy observations
Kalman estimate
95% band
True signal, hidden from the filter

How noisy the filter thinks each measurement is trust datatrust model R = 0.36, the true value

Simulated random walk, seed 1606. Hover or drag across the plot to read any step. One measurement at a time, a little less guesswork.

Fig. 2 · False discovery rate How easily p-values fool us

Scoreboard
Rule	Flagged	Noise	FDP
p < 0.05	–	–	–
BH, q = 0.05	–	–	–

Flagged
Real effect
False discovery
Missed real effect

Simulated: 100 independent z-tests, 10 with a real effect (μ = 3), seed 1234. FDP is the share of flagged tests that are noise in this one draw. BH holds its average, the FDR, at q. Hit New draw and watch p < 0.05 fool you differently every time.

Peer-reviewed publications

Ongoing research projects

01 / Peer-reviewed publications

Published research

Two peer-reviewed articles in Food Research International (2026).

Co-first author Food Res. Int.

Lineage-aware comparative genomics shows no consistent reservoir-specific plasmidome in low-moisture food-associated Salmonella Enteritidis

Tenzin, K.*, Balyatanda, S. B.*, Doddabematti Prakash, S., Siliveru, K., & Platt, T. G. (2026).

*Equal contribution

Food Research International, 242, 120003 · 2,824 records curated · 51 SNP clusters · 10 mixed-reservoir clusters · 64 genomes analyzed

doi.org/10.1016/j.foodres.2026.120003

Statistical evidence

H₀: OR_MH = 1

IncFIB(S)

OR: 0.93
FDR-adjusted p: 0.948

IncFII(S)

OR: 1.59
FDR-adjusted p: 0.797

IncX1

OR: 0.64
FDR-adjusted p: 0.797

No replicon family showed statistically significant reservoir-associated enrichment after lineage stratification and FDR correction. Nonsignificance does not establish the absence of an effect.

Explore this study

A lineage-aware view of the plasmidome

Explore published aggregate results for Salmonella Enteritidis from low-moisture food (LMF) and non-LMF sources. Controls change the view of reported results; they do not run new statistical tests or filter individual genomes.

From curated records to comparative genomes

Select a stage to see why it matters. Grouping records into lineages, selecting assemblies, and applying QC are distinct steps.

Curated metadata, not a sequencing experiment

The study curated 2,824 public non-clinical records: 45 LMF and 2,779 non-LMF. This is the metadata pool; not every record entered assembly-level analyses, and these genomes were not newly sequenced for this study.

Section 3.1, p. 5

Distinct metadata and assembly stages · Sections 2.2 and 3.1–3.2
Stage	LMF	Non-LMF	Total
Curated records	45	2,779	2,824
Records in mixed clusters	24	319	343
Selected before QC	24	41	65
Final assembly set	24	40	64

Replicon prevalence within the final genome set

Inspect a replicon

Genomes carrying each replicon

LMF · n = 24Non-LMF · n = 40

Counts for all families are listed in the table below.

Select a chart group or use the dropdown. Bar lengths show descriptive prevalence, not adjusted effect estimates.

IncFIB(S)

LMF: 22 / 24 (91.7%) · Non-LMF: 36 / 40 (90.0%).

Reported CMH odds ratio: 0.93
Reported FDR-adjusted p: 0.948

No statistically significant reservoir association after the reported FDR correction.

CMH estimates account for SNP-cluster membership. They are taken from the paper, not recalculated from the pooled bars. OR = 1 is the reference for no association. Confidence intervals are not plotted.

Section 3.3, pp. 5–6 · Fig. 4 · FDR threshold: 0.05
Family	LMF	Non-LMF	CMH OR	FDR p
IncFIB(S)	22 / 24	36 / 40	0.93	0.948
IncFII(S)	22 / 24	35 / 40	1.59	0.797
IncX1	4 / 24	10 / 40	0.64	0.797

Genome-level predicted plasmid mobility

Each bar represents genomes carrying at least one reconstructed plasmid in that predicted class. A genome may contribute to more than one class.

Inspect a mobility class

Genomes with at least one plasmid in each class

LMF · n = 24Non-LMF · n = 40

Counts for all classes are listed in the table below.

The classes overlap; percentages are not parts of a single whole.

Conjugative

16.7% / 35.0%

LMF: 4 / 24 · Non-LMF: 14 / 40.

Predicted conjugative potential is based on MOB-suite typing. It is not an experimental measurement of transfer.

The genome with no reconstructed plasmid remains in the denominators.

Section 3.4, pp. 6–7 · Fig. 5B
Class	LMF	Non-LMF
Conjugative	4 / 24	14 / 40
Mobilizable	4 / 24	6 / 40
Non-mobilizable	23 / 24	33 / 40

Gene-cluster composition

Explore frequency classes in the full pangenome, then switch to reservoir sharing among the accessory clusters.

Full pangenome · 4,899 gene clusters

Both compositions are listed in the tables below.

Core

4,026 / 4,899 · 82.2%

Present in at least 99% of the analyzed genomes, using the paper's frequency definition.

LMF-only and non-LMF-only describe occurrence in this sampled dataset. No accessory gene cluster showed significant LMF-associated enrichment after the reported FDR correction.

Full pangenome · Section 3.6, p. 9 · Fig. 7
Category	Gene clusters	Genome-frequency definition
Core	4,026	≥99%
Soft core	167	95% to <99%
Shell	297	15% to <95%
Cloud	409	<15%

Accessory occurrence · Section 3.6, pp. 9–10
Occurrence in this dataset	Accessory clusters
Shared between reservoirs	707
LMF-only	55
Non-LMF-only	111

What the evidence supports

Population structure matters: replicon and plasmid profiles were largely lineage-associated. No tested replicon family or accessory gene cluster showed statistically significant LMF-associated enrichment after the reported FDR correction. Select a point to explore its meaning.

Compare within genetic backgrounds

Shared ancestry can confound pooled reservoir comparisons. The study restricted comparisons to mixed SNP clusters and used lineage-stratified association tests; plasmid variation was largely associated with lineage.

Sections 2.4 and 4.1–4.2

These are observational genomic comparisons, not experimental demonstrations of adaptation.

This explorer summarizes published aggregate findings from Tenzin et al., Food Research International 242 (2026), 120003. Sources: Sections 2.2, 2.4, 3.1–3.4, 3.6 and 4.1–4.3; Figs. 1–5 and 7. Displayed percentages are calculated from the reported counts. DOI: 10.1016/j.foodres.2026.120003.

Co-author Food Res. Int.

Benchmarking a 16S rRNA sequencing protocol for microbiome analysis in low-moisture grain environments

Doddabematti Prakash, S., Balyatanda, S. B., Sytsma, J., Tenzin, K., & Siliveru, K. (2026).

Food Research International, 241, 119710 · 32 pipeline–database combinations compared

doi.org/10.1016/j.foodres.2026.119710

02 / Ongoing work

Ongoing research

Two manuscripts in preparation and two developing projects, spanning microbial evolution, genome resources, and scientific computing.

Manuscript in preparation · Project lead

Population genomicsPlant-associated

Agrobacterium on wild sunflowers

Kansas State University · Platt Lab

How do closely related bacterial groups vary within the same plant habitat? I am investigating genomic diversity and evolutionary relationships in wild sunflower-associated Agrobacterium.

In notation

\begin{gathered}\text{core}=\bigcap_{i=1}^{n} G_i\\[4pt] \text{accessory}=\Big(\bigcup_{i=1}^{n} G_i\Big)\setminus\text{core}\end{gathered}

$G_i$ is the set of gene families in strain $i$ . The core is what every strain carries. The accessory part is what only some carry, and it is where mobile elements tend to live.

d_J(i,j)=1-\frac{|G_i\cap G_j|}{|G_i\cup G_j|}

Gene-content distance between two strains. Two strains can sit close together on the tree and still be far apart here.

View approach

The work combines genome quality assessment, comparative genomics, and phylogenetic analysis to examine relationships among plant-associated bacteria. Understanding this variation can inform broader questions about microbial diversity in shared habitats. The manuscript is in preparation.

Manuscript in preparation · First author

Annotation integrationGenome resources

Integrating Volvox genome annotations

Kansas State University · Olson Lab

How can successive genome releases be compared without mistaking annotation differences for biological change? I am developing approaches to integrate Volvox carteri gene annotations and evaluate their consistency.

In notation

\begin{gathered}\mathcal{B}=(A\cup B,\;E)\\[2pt] E=\{(a,b)\in A\times B:\ a\sim b\}\end{gathered}

$A$ and $B$ are the gene models from two genome releases. An edge $a\sim b$ means sequence comparison supports them as the same locus.

\operatorname{type}(K)=\begin{cases}\text{one-to-one} & |K_A|=|K_B|=1\\ \text{split} & |K_A|=1<|K_B|\\ \text{merge} & |K_A|>1=|K_B|\\ \text{complex} & |K_A|,\,|K_B|>1\\ \text{orphan} & \min(|K_A|,|K_B|)=0\end{cases}

Each connected component $K$ of $\mathcal{B}$ , with $K_A=K\cap A$ , gets one label. Splits, merges, and orphans are where an annotation change can masquerade as biology.

View approach

The approach brings together sequence comparisons, annotation reconciliation, and genomic-context analysis, with conserved-gene and RNA-seq evidence used to assess candidate gene models. The goal is to support more reliable comparative research in a model of multicellularity. A first-author manuscript is in preparation.

Active research project · External validation

Pathogen genomicsFood safety

Evaluating Vibrio markers in seafood

Comparative genomics · Vibrio vulnificus

Do published clinical-associated markers remain informative when evaluated against seafood isolates? I am investigating their transferability across sampling contexts, with food-safety surveillance as the motivation.

In notation

\pi_m(s\mid c)=\Pr\big(X_m=1\mid S=s,\ C=c\big)

$X_m=1$ when an isolate carries published marker $m$ . $S$ is its sampling source, clinical or seafood, and $C$ is its clade.

\Delta_m(c)=\pi_m(\text{clinical}\mid c)-\pi_m(\text{seafood}\mid c)

Compared within clades, not pooled. Pooled, a marker can look clinical just because clinical isolates come from a few lineages. Within a clade, the marker has to carry the signal itself.

View approach

The project uses curated public genome metadata, genome quality assessment, and phylogeny-aware comparisons to evaluate published marker candidates in a different sampling context. The aim is to assess the evidence needed before applying such markers to food-safety surveillance. Evaluation is ongoing.

Also developing: whether AI agents can help researchers assess if a conclusion in comparative bacterial genomics is supported by its evidence.

03 / Research experience

Experience

Graduate Researcher

Kansas State University · Platt Lab

Jun 2025 – present

Pathogen and plant-associated population genomics, 16S benchmarking, and reproducible HPC workflows.

Read contributions

Lineage-aware pathogen genomics. Co-led a study of Salmonella Enteritidis from low-moisture foods, integrating NCBI Pathogen Detection metadata, SNP-cluster stratification, plasmid and antimicrobial resistance analyses, and pangenomics. The analysis found no consistent reservoir-specific plasmidome after accounting for lineage; published as co-first author in Food Research International (2026).
Methodological benchmarking. Contributed to conceptualization, methodology, visualization, and manuscript review and editing for a benchmark of 32 bioinformatics pipeline–database combinations for 16S rRNA microbiome analysis of low-moisture grain environments, in collaboration with K-State’s Department of Grain and Food Science.
Independent project leadership. Leading comparative population-genomics research on plant-associated Agrobacterium, examining genomic diversity and evolutionary relationships.
Reproducible genomics. Developed repeatable workflows for genome quality control, species identification, and metadata harmonization using Slurm and Apptainer, with provenance records.

Graduate Researcher

Kansas State University · Olson Lab

Aug 2023 – Jun 2025

Genome-version integration and evolutionary transcriptomics in volvocine algae.

Read contributions

Genome-version integration. Developed approaches for integrating Volvox carteri genome annotations using sequence comparisons and annotation metadata; a first-author manuscript is in preparation.
Evolutionary transcriptomics. Analyzed Pandorina differential expression and co-expression patterns, and comparative Gonium/Pandorina developmental data using clustering and multidimensional scaling.

Research Staff · GeoVista & DiNonA

Kumho HT, Seoul

Jan 2021 – Jul 2023

Antibody drug development (DiNonA) and biopharmaceutical consulting (GeoVista): investigational product documentation, preclinical data analysis, and Korean–English partner support.

Read contributions

DiNonA · Research and Development. Managed investigational product documentation and analysis for antibody development programs. Translated Korean–English investigational new drug (IND) application materials, chemistry, manufacturing and controls (CMC) documentation, and standard operating procedures (SOPs).
GeoVista · Consultancy. Evaluated preclinical models and analyzed therapeutic research data; supported partner negotiations as a Korean–English interpreter and translator. Supported business development and communication with international partners. Presented preclinical study plans and results to international CRO partners.

Regulatory / CMCIND / SOPKorean-English translation

Earlier research training

Project Intern

Korea University College of Medicine, Seoul

Aug – Dec 2020

Generated GBA1 knockout cell lines with CRISPR/Cas9 and expressed wild-type and mutant GBA1 for Gaucher disease research; assessed expression by immunoblotting.

CRISPR/Cas9ImmunoblottingCell culture

U-WURF Research Fellow

Ulsan National Institute of Science and Technology (UNIST), Ulsan

Cloned SpyTag/SpyCatcher constructs to load cargo into encapsulin nanocages; purified encapsulins by sucrose-gradient ultracentrifugation and size-exclusion chromatography.

EncapsulinsProtein purificationMolecular cloning

04 / Education

Education

Ph.D. in Biology

Kansas State University

Aug 2023 – May 2028
expected

Manhattan, Kansas · Platt Lab. Lineage-aware pathogen genomics, food-safety surveillance, and plant-associated population genomics; two manuscripts in preparation and two developing projects.

PLATT LAB2 MANUSCRIPTS IN PREPINSTRUCTOR OF RECORD

B.Sc. in Life Science

University of Seoul

Feb 2016 – Feb 2020

Seoul, South Korea · Completed through Korean-medium instruction. Funded by the Korean Government Scholarship for Undergraduate Studies, covering five years of full support, including one year of Korean language training. Recipient of the Korean Language Excellence Grant.

KOREAN GOV'T SCHOLARSHIPLANGUAGE EXCELLENCE GRANT

05 / Presentations

Presentations

Four presentations in 2026.

2026July

Accessory gene content differences in low-moisture food-associated Salmonella Enteritidis: a Pathogen Detection, SNP cluster-stratified pangenome analysis

Poster · International Association for Food Protection (IAFP) Annual Meeting, New Orleans, LA · July 2026

Poster

2026March–May

Lineage stability overrides reservoir adaptation in the plasmidome of Salmonella Enteritidis from low-moisture foods

Flash talk · 5th Annual KU Center for Genomics Symposium, Lawrence, KS · May 2026

Oral · K-State Graduate Research, Arts, and Discovery Forum, Manhattan, KS · April 2026

Oral · 51st Annual Biology Graduate Student Research Forum, K-State, Manhattan, KS · March 2026

Flash talk / Oral

06 / Honors

Awards & fellowships

2026

Robert J. Robel Graduate Student Award in Wildlife Biology and Ecology

Division of Biology, Kansas State University. Recognizes research contributions in wildlife biology and ecology.

2026

IAFP Student AI Benchmark Competition · Finalist

Team-based prediction of Listeria in soil, International Association for Food Protection.

2026 – 2027

International Student Scholarship

K-State Alumni Association.

2026 – 2027

Risk Management Fellow

K-State Risk Management Center: interdisciplinary student teams working on risk projects with industry partners.

Ongoing

Research travel & training awards

K-State Graduate Student Council, College of Arts and Sciences, and Biology Graduate Student Association.

2016 – 2020

Korean Government Scholarship for Undergraduate Studies

Supported undergraduate studies at the University of Seoul; also received the Korean Language Excellence Grant.

07 / Teaching & service

Teaching & service

Teaching and student leadership.

Instructor of Record

BIOL 632 · Ecological Analysis Laboratory

Spring 2026. Taught experimental design, R-based data analysis, scientific writing, peer review, and conference-style oral presentations.

Graduate TA

BIOL 201 & BIOL 198

2023 – 2025. Led organismal biology laboratories and supported introductory genetics instruction.

Leadership

Student leadership

Vice President, Biology Graduate Student Association · 2026 – 2027. Help organize graduate orientation, peer support, and student events.

Vice President, International Coordinating Council · 2024 – 2025. Launched the council's first podcast, addressing questions international students hesitate to ask publicly; organized a talent show (200+ attendees) and sporting events (100+ participants).

Community

Earlier service

Founding President, Bhutanese Association in South Korea. Founded the association, built its website, raised funds, and organized community events.

Board Member, International Student Association, University of Seoul.

08 / Methods

Methods & tools

My loadout: the tools I actually reach for.

1Comparative genomicsNCBI Pathogen DetectionPanarooPPanGGOLiNMOB-suitePlasmidFinderAMRFinderPlusIQ-TREEGubbins
2Assembly & QCSPAdesFlyeMedakaUnicyclerCheckM2QUAST
3Annotation & RNA-seqBaktagffutilsSTARnf-core/rnaseq
4Workflows & HPCNextflownf-coreSnakemakeApptainerDockerSlurm
5StatisticsRPythonCMH stratified testsFDR controlPERMANOVA
6Wet labCRISPR/Cas9Molecular cloningProtein purification
7LanguagesDzongkha and Sharshopa (native)EnglishKorean (TOPIK 5)NepaliHindi

09 / Contact

Get in touch.

Interested in microbial genomics, food safety, or reproducible workflows? I welcome research collaborations. Based at Kansas State University in Manhattan, Kansas.

Emailtenzin12@ksu.edu

ORCID0009-0009-2417-6880

GitHubtkinley