AI features now support more model providers including OpenAI, Anthropic, Gemini, OpenRouter, and OpenAI-compatible models. The default model is still deepseek-chat (#15).
polish_report() with use_ai = TRUE now includes a summary of the analysis results at the end of the report, generated by the LLM using multimodal reasoning (#16).
Minor improvements and bug fixes
check_glysmith_deps() now checks only the packages required by the supplied blueprint, instead of asking users to install every optional dependency. Step-specific glycoverse packages are optional again, with installation guidance for the glycoverse r-universe repository (#17).
Enrichment analysis now uses glyfun instead of glystats, as enrichment analysis functions in glystats have been deprecated in favor of glyfun (#10).
quench_result() no longer leaves a stray Rplots.pdf in the working directory (#18).
glysmith 0.10.1
Minor improvements and bug fixes
step_preprocess() removes QC samples after preprocessing as documented.
glysmith 0.10.0
Breaking Changes
Split step_quantify_motifs() into two separate functions:
Updated all downstream steps to accept the new data types (dynamic_motif_exp, sig_dynamic_motif_exp, branch_motif_exp, sig_branch_motif_exp) instead of the old motif_exp/sig_motif_exp.
Differentiated DEA results: dynamic_motif_exp now generates dynamic_dma_res and branch_motif_exp generates branch_dma_res.
Minor improvements and bug fixes
Add a vignette about creating a blueprint manually.
glysmith 0.9.1
Minor improvements and bug fixes
Update dependency strategy to use the r-universe repo.
glysmith 0.9.0
New features
Add loadings and screeplot parameters to step_pca() for optional generation of loading plots and scree plots.
Add check_glysmith_deps() for checking and installing optional dependencies required by blueprint steps.
forge_analysis() now checks if all dependencies are installed before the actual analysis.
Minor improvements and bug fixes
Add ReactomePA to Suggests for pathway enrichment analysis.
glysmith 0.8.1
Minor improvements and bug fixes
Fix a bug in step_quantify_motifs() that bisecting GlcNAc was regarded as a branching GlcNAc.
Add cast_data() to fetch data from glysmith_result object.
When the name parameter is not specified in cast_plot(), cast_table(), or cast_data(), the functions return the names of all available objects. This makes it easier to know what objects are available.
inquire_blueprint() now prints a brief description of the blueprint to console, instead of an explanation of why the steps are chosen. There is a slight difference in the wording.
Optimize the prompt of inquire_blueprint() to let it group steps using br(). Also, it now understands all step arguments and can use them if needed.
“Tables” section is removed from the report. It is duplicated with the README.md file generated by quench_result().
Add a new vignette about AI features.
glysmith 0.4.0
Breaking changes
The dynamic argument mechanism was removed. Now you can directly pass arguments to the step functions.
New features
Added br() for creating branches in a blueprint. inquire_blueprint() also supports blueprints with branches now.
step_dea_limma() now support covariates and paired comparison, via the covariate_cols and subject_col arguments.
Minor improvements and bug fixes
Fix the issue that blank lines were printed to console when the package is first loaded in one session.
Update system prompt and the reflection strategy in inquire_blueprint() to make the LLM work better.
Moved several dependencies (pROC, Rtsne, uwot, EnhancedVolcano, org.Hs.eg.db, clusterProfiler, ggplotify, pheatmap, and factoextra) to “Suggests”. These are now automatically checked and installed when required by blueprint steps.
glysmith 0.2.0
New features
Add inquire_blueprint() use a Large Language Model to generate a blueprint from user prompt.
Add step_tsne() for t-distributed Stochastic Neighbor Embedding.
Add step_umap() for Uniform Manifold Approximation and Projection.
Add step_roc() for Receiver Operating Characteristic analysis.
step_pca() now also generates loading plot and screeplot, besides the individual plot.
step_pca() (along with the newly added step_tsne() and step_umap()) now has an on parameter to control which experiment to perform the analysis on.
