process_controversial_clusters

Process controversial clusters through discussion

internal

Automated cell type annotation for single-cell RNA sequencing data
using consensus predictions from multiple large language models (LLMs). LLMs are
artificial intelligence models trained on vast text corpora to understand and
generate human-like text. This package integrates with 'Seurat' objects and
provides uncertainty quantification for annotations. Supports various LLM
providers including 'OpenAI', 'Anthropic', and 'Google'. The package leverages
these models through their respective APIs (Application Programming Interfaces)
<https://platform.openai.com/docs>, <https://docs.anthropic.com/>, and
<https://ai.google.dev/gemini-api/docs>.
For details see Yang et al. (2025) <doi:10.1101/2025.04.10.647852>.

Chen Yang

mLLMCelltype

Cell Type Annotation Using Large Language Models

process_controversial_clusters function

<dl><dt>controversial_clusters</dt>
<dd>List of controversial cluster IDs</dd>
<dt>input</dt>
<dd>Either the differential gene table or a list of genes</dd>
<dt>tissue_name</dt>
<dd>The tissue type or cell source</dd>
<dt>successful_models</dt>
<dd>Vector of successful model names</dd>
<dt>api_keys</dt>
<dd>Named list of API keys</dd>
<dt>individual_predictions</dt>
<dd>List of predictions from each model</dd>
<dt>top_gene_count</dt>
<dd>Number of top differential genes to use</dd>
<dt>controversy_threshold</dt>
<dd>Threshold for marking clusters as controversial</dd>
<dt>max_discussion_rounds</dt>
<dd>Maximum number of discussion rounds for controversial clusters</dd>
<dt>cache_manager</dt>
<dd>Cache manager object</dd>
<dt>use_cache</dt>
<dd>Whether to use cached results</dd>
<dt>consensus_check_model</dt>
<dd>Model to use for consensus checking</dd>
<dt>force_rerun</dt>
<dd>Whether to force re-analysis, ignoring cache</dd></dl>

Arguments

Process controversial clusters through discussion — process_controversial_clusters

<dl>

<dt>controversial_clusters</dt>
<dd>List of controversial cluster IDs</dd>


<dt>input</dt>
<dd>Either the differential gene table or a list of genes</dd>


<dt>tissue_name</dt>
<dd>The tissue type or cell source</dd>


<dt>successful_models</dt>
<dd>Vector of successful model names</dd>


<dt>api_keys</dt>
<dd>Named list of API keys</dd>


<dt>individual_predictions</dt>
<dd>List of predictions from each model</dd>


<dt>top_gene_count</dt>
<dd>Number of top differential genes to use</dd>


<dt>controversy_threshold</dt>
<dd>Threshold for marking clusters as controversial</dd>


<dt>max_discussion_rounds</dt>
<dd>Maximum number of discussion rounds for controversial clusters</dd>


<dt>cache_manager</dt>
<dd>Cache manager object</dd>


<dt>use_cache</dt>
<dd>Whether to use cached results</dd>


<dt>consensus_check_model</dt>
<dd>Model to use for consensus checking</dd>


<dt>force_rerun</dt>
<dd>Whether to force re-analysis, ignoring cache</dd>

</dl>

process_controversial_clusters: Process controversial clusters through discussion

Description

Usage

Value

Arguments