parse_text_predictions

Handles multiple output formats from LLMs:<ul>
<li>"cluster_id: cell_type" format</li>
<li>"1. cell_type" numeric index format</li>
<li>Positional fallback (line index maps to cluster index)</li>
</ul>

internal

Automated cell type annotation for single-cell RNA sequencing data
using consensus predictions from multiple large language models. Integrates
with Seurat objects and provides uncertainty quantification for annotations.
Supports various LLM providers including OpenAI, Anthropic, and Google.
For details see Yang et al. (2025) <doi:10.1101/2025.04.10.647852>.

Chen Yang

mLLMCelltype

Cell Type Annotation Using Large Language Models

parse_text_predictions function

<dl><dt>model_preds</dt>
<dd>Character vector of prediction lines from a model</dd>
<dt>all_clusters</dt>
<dd>Optional character vector of cluster IDs for positional fallback</dd></dl>

Arguments

Parse text-format model predictions into a named list — parse_text_predictions

<dl>

<dt>model_preds</dt>
<dd>Character vector of prediction lines from a model</dd>


<dt>all_clusters</dt>
<dd>Optional character vector of cluster IDs for positional fallback</dd>

</dl>

parse_text_predictions: Parse text-format model predictions into a named list

Description

Usage

Value

Arguments