How to verify gene model annotations?

Understand Gene Models

• Review the biological context of the gene and its family to predict possible gene structure complexity.

 

• Understand the organism's genome annotation history which can suggest common annotation issues in specific genomes.

Collect Supporting Evidence

• Compile RNA-Seq data for transcript evidence supporting the gene model prediction.

 

• Gather EST (Expressed Sequence Tag) data to validate regions of transcription.

 

• Use protein alignment data to check if predicted exons align well with known protein sequences.

Use Comparative Genomics

• Identify orthologous genes in related species and analyze conserved regions to infer possible exon-intron structures.

 

• Compare synteny blocks to preserve gene order and orientation, validating gene model consistency across species.

Employ Computational Tools

• Use gene prediction software tools like Augustus or GeneMark to provide independent verification of the gene model.

 

• Deploy splice site prediction tools to review the accuracy of intron-exon boundaries.

Manually Curate Annotations

• Manually inspect gene models in genome browsers like UCSC Genome Browser or Ensembl to visually confirm exon structures and transcript alignment.

 

• Refine models to resolve discrepancies using supporting evidence gathered from experimental data and comparative analysis.

Validate with Experimental Data

• Use RT-PCR analysis to experimentally validate predicted exon combinations and intron boundaries.

 

• Perform CRISPR/Cas9 gene editing to confirm functional relevance of predicted gene models by observing phenotypic consequences of edits.

Document Findings

• Record all annotation decisions, including the data sources and rationale used in adjustments, for tractability and future reference.

 

• Update public databases or report findings to genome annotation community to enhance collective data accuracy.