Photo © PDS Lab.
SmarLens
Beta v2.1.0SmarLens is a milk thistle genomics workspace designed to move from gene discovery to functional interpretation and experimental design. Browse the genome, search Smar genes or DNA sequences, compare Arabidopsis-informed homologs, run GO analysis, prioritize silymarin pathway candidates, and design CRISPR guides or PCR primers in one focused beta platform.
Photo © PDS Lab.
Browse
Search
Analyze
About SmarLens
SmarLens is a beta-stage genomics workspace for gene-centered exploration and analysis of milk thistle (Silybum marianum).
Disclaimer
SmarLens is provided as a research-oriented beta resource. The developers and affiliated laboratory do not warrant, and do not assume legal liability or responsibility for, the accuracy, completeness, or usefulness of any data, annotation, analysis result, software output, document, or related information made available through this tool. All computational results should be interpreted as supporting evidence and require independent biological and experimental validation.
Purpose
SmarLens was developed to make milk thistle genome resources practical for gene-level discovery, interpretation, and experimental planning. The current beta brings genome browsing, Smar gene reports, DNA sequence search, Arabidopsis-informed homology, GO enrichment, silymarin candidate prioritization, CRISPR guide design, and PCR primer design into one connected workspace.
Reference Genome and Annotation
The core value of SmarLens depends directly on the chromosome-level milk thistle genome assembly and gene annotation. Without the gene models, transcript coordinates, protein sequences, and feature annotations generated from this reference resource, SmarLens could not provide gene reports, genome viewer tracks, sequence extraction, GO inference, primer design, guide RNA context, or candidate prioritization.
The genome assembly and annotation used by SmarLens are based on Kim et al., 2024. Public source files are available from the associated figshare dataset. SmarLens development is being carried out in close connection with the annotation work, and the annotation resource should be cited as the primary biological foundation of this database.
Implementation and Attribution
SmarLens is an integration and interface layer built around public genome resources, established bioinformatics software, and open-source logic where appropriate. The project does not claim to have invented the underlying alignment, primer design, guide RNA, genome browsing, domain search, or ontology algorithms. Its contribution is the milk thistle-specific data organization, workflow integration, visualization, caching, scoring framework, and user-facing implementation.
Genome browsing
Genome visualization is implemented with JBrowse 2 using SmarLens-prepared reference and annotation tracks.
Similarity, domains, and trees
Protein homology uses DIAMOND with BLASTP fallback. DNA search uses local NCBI BLAST+ databases. Protein domains and family views use Pfam/HMMER, MAFFT, and FastTree.
CRISPR guide RNA design
The guide RNA workflow follows the same target/PAM and off-target counting principle as CRISPRdirect, with milk thistle-specific jellyfish exact-count indexes and SmarLens-specific result presentation.
PCR primer design
Primer candidates are generated with Primer3, then annotated with milk thistle genome specificity checks.
Functional inference
GO analysis is inferred through Arabidopsis homolog relationships and Gene Ontology annotations. These results should be treated as homology-supported functional hypotheses, not curated milk thistle experimental annotations.
Silymarin prioritization
Silymarin Prioritizer ranks Smar genes using pathway keyword evidence, inferred GO evidence, Arabidopsis homology, flower-biased expression, gene model quality, and local pathway-neighborhood context. The score is intended for hypothesis prioritization, not as proof of pathway function.
Current Scope
SmarLens v2.1.0 is organized into three working areas: Browse for genome viewing and BLAST-like sequence search, Search for Smar gene reports and Arabidopsis-to-Smar similarity analysis, and Analyze for GO enrichment, silymarin candidate prioritization, CRISPR guide RNA design, and PCR primer design. Current data links and method notes are collected in the Resources page.
Public Test Limits
For public beta testing, heavy protein homology analyses such as Find Similar Gene are limited to 10 input queries per request. Gene Search, GO analysis, BLAST Search, guide RNA design, primer design, and genome browsing use task-specific limits such as input size, result count, sequence length, or genomic span to protect server availability.
Development Team
Janghyun Choi, Ph.D.
Inha University · Department of Biological Sciences
Main developer and code architect. Designed and implemented the SmarLens interface, pipeline modules, scoring workflows, and integrated analysis components.
Sinhoe Ahn M.S. candidate
Inha University · Department of Biological Sciences
Contributed to module implementation, pipeline testing, data processing, logic refinement, and validation of analysis outputs.
Jinmi Yoon, Ph.D.
Inha University · Department of Biological Sciences
Kyung Do Kim, Ph.D.
Myongji University · Biosciences and Bioinformatics
Lae-Hyeon Cho, Ph.D.
Pusan National University · Plant Bioscience
Provided project direction, biological interpretation, genome resource guidance, funding support, and overall supervision for database development.
Maintenance
Pipeline maintained by Janghyun Choi · jchoi@inha.ac.kr
Source Code
Source code and issue tracking are available on GitHub.
Resources
Reference data, downloadable files, and concise method notes for the current SmarLens beta.
Data Availability
SmarLens is built around the chromosome-level milk thistle genome assembly, gene annotation, transcript/protein sequences, expression summaries, and locally prepared analysis resources. The downloads below are data files only; BLAST, DIAMOND, jellyfish, tabix, and other index files are not listed for direct download.
Downloadable Data
- SmarLens SQLite database Source: generated in this project. Integrated gene models, expression summaries, homology, GO, known Silybum matches, and cached analysis tables.
- Milk thistle genome FASTA Source: Kim et al., 2024 reference dataset. Chromosome-level Silybum marianum EM05 genome sequence.
- Milk thistle gene models GFF Source: Kim et al., 2024 reference dataset. Gene, transcript, exon, and CDS annotation used by SmarLens.
- Milk thistle protein FASTA Source: Kim et al., 2024 reference dataset. Predicted Smar protein sequences from the published annotation.
- Milk thistle expression count table Source: processed in this project from the Kim et al., 2024 EM05 RNA-seq dataset, BioProject PRJNA1021369. The file is a seven-sample count matrix, not a DEG result table.
- Arabidopsis protein FASTA Source: Ensembl Plants TAIR10 peptide FASTA. Used as the Arabidopsis protein reference for homology workflows.
- Reference genome dataset
- Arabidopsis TAIR10 annotation source
- Reference publication
- Source code on GitHub
CDS/transcript FASTA exports and additional table-level exports can be added after the final release checks. A formal SmarLens citation will be added when available.
Local Analysis Resources
Genome viewer: JBrowse 2 track files are served from SmarLens-prepared reference and annotation files.
Sequence search and homology: local BLAST databases and DIAMOND indexes support DNA sequence search and Arabidopsis-informed protein similarity workflows.
Guide RNA and primer design: jellyfish exact-count indexes and local genome specificity checks are used to summarize candidate uniqueness in the milk thistle reference genome.
Functional analysis: GO enrichment and silymarin candidate prioritization use Arabidopsis homolog-inferred annotations, expression evidence, protein/domain evidence, and milk thistle gene model context.
Use Notes
SmarLens is a beta-stage research tool. Homology, GO, CRISPR, primer, BLAST, and prioritizer outputs should be treated as computational evidence and interpreted together with independent biological validation.
Search Gene
Silybum marianum gene lookup
Find Similar Gene
Identify milk thistle genes similar to known Arabidopsis genes.
- Resolve each Arabidopsis symbol or AT gene ID to protein isoforms, then let you keep only the isoforms to compare.
- Search selected Arabidopsis proteins against milk thistle proteins using DIAMOND blastp, with BLASTP fallback when needed.
- Rank Smar candidates by protein similarity, then calculate shared Pfam-domain identity with MAFFT for detailed pairwise inspection.
- Use results as homology evidence, not proof of orthology or conserved function; validate key candidates with domain, GO, expression, and experimental context.
BLAST Search
Find milk thistle genomic loci and Smar genes from pasted DNA sequence.
Paste a FASTA or plain DNA sequence. SmarLens searches the milk thistle reference genome with local NCBI BLASTN, filters hits by identity and query coverage, then annotates each locus with overlapping Smar genes or the nearest Smar gene when no overlap exists. Short queries use the BLASTN short-task setting; longer queries use standard BLASTN.
Guide RNA Design
Design candidate CRISPR/Cas9 guide RNAs from a milk thistle gene region or pasted DNA sequence.
SmarLens generates guide RNA candidates using the same PAM-scanning principle as CRISPRdirect: it scans both strands of the input DNA sequence, detects the selected PAM motif, and extracts the adjacent 20 nt protospacer as a guide candidate. Milk thistle-specific specificity checks are then performed locally against the indexed Silybum marianum reference genome, and the results are displayed in a SmarLens-oriented format.
PCR Primer Design
Design PCR primer pairs from a milk thistle gene region or pasted DNA sequence.
SmarLens uses Primer3-compatible constraints for primer length, Tm, GC%, and amplicon size, then checks milk thistle genome specificity at both the individual primer level and the primer-pair product level. If primer3_core is installed on the server, SmarLens uses it; otherwise it falls back to a local constraint-based primer scanner.
Genome Viewer
Browse milk thistle chromosomes, genes, transcripts, exons, and CDS features in an embedded JBrowse 2 genome browser.
Silymarin Prioritizer
Rank milk thistle genes by weighted pathway, GO, homology, expression, gene model, and neighborhood evidence for silymarin-related hypotheses.
This tool ranks milk thistle genes as silymarin-related candidates by combining interpretable computational evidence: pathway-related keyword support from Arabidopsis homolog descriptions, inferred GO/domain hints, homolog confidence, flower-biased RNA-seq expression, gene model completeness, and local candidate-gene neighborhood. The score is intended for hypothesis prioritization only and should not be interpreted as experimental validation of silymarin biosynthetic function.
Functional Analysis
Run Arabidopsis homolog-inferred GO enrichment for milk thistle gene sets.
SmarLens maps milk thistle genes to Arabidopsis homologs, transfers Arabidopsis GO terms as inferred annotations, and tests whether submitted Smar genes are over-represented for each GO term using a one-sided hypergeometric test. The default background is all Smar genes with inferred GO annotation under the selected filters; an optional custom background can be used for RNA-seq or other experiments where only a subset of genes was testable. Results are functional hypotheses and should be interpreted with homolog confidence and biological context.