---
title: BioBERT inputs & parameters
description: Research query, max results, similarity threshold and entity-type options, with 512-token segments. BioBERT is deprecated and chat cannot call it.
---

You provide a research query or biomedical text — typed, uploaded, or fetched from PubMed — along with the task and entity types to extract. The similarity threshold most shapes what comes back: it sets the relevance cutoff that decides which papers and entities are kept.

> **Not available in chat**

> BioBERT is deprecated on the platform and not registered with the orchestrator, so chat cannot call it. A current LLM performs BioBERT's entity extraction and relevance ranking natively over text retrieved with [PubMed](/docs/pubmed); for curated identifiers an LLM must not invent (NCBI Gene IDs, MeSH terms), use PubTator3. Existing BioBERT job results remain viewable.

## Webapp input methods

The .bloom* webapp provides a streamlined interface for BioBERT literature search.

## Query input

- **Label**: "Research Query"
- **Placeholder**: "Enter your research question or keywords..."
- **Character limit**: Displayed as `{current}/{max}` counter
- **Example queries** (click to use):
  - "CRISPR gene editing COVID-19 treatment"
  - "Alzheimer's disease tau protein aggregation"
  - "mRNA vaccine immunogenicity mechanisms"
  - "CAR-T cell therapy cancer treatment"
  - "Machine learning drug discovery"

## Webapp parameters

### Basic settings

| Parameter | Range | Default | Description |
|-----------|-------|---------|-------------|
| **Max Results** | 1-500 | 100 | Papers to retrieve |
| **Similarity Threshold** | 0-1 | 0.7 | Relevance cutoff (slider with 0.5/0.7/0.9 marks) |

### Analysis options (checkboxes)

| Option | Default | Description |
|--------|---------|-------------|
| **Semantic Analysis** | Enabled | AI-powered content understanding |
| **Extract Biomedical Entities** | Enabled | Genes, proteins, diseases, chemicals |
| **Generate AI Summaries** | Enabled | Automated abstract summarization |
| **Citation Analysis** | Disabled | Analyze citation networks |

### Advanced filters (accordion panel)

| Parameter | Description |
|-----------|-------------|
| **Date From** | Filter papers published on or after date |
| **Date To** | Filter papers published on or before date |
| **Include Full Abstracts** | Include complete abstract text (checkbox, default: enabled) |

## Submit button

- **Label**: "Start Literature Search"
- **Disabled when**: Query is empty
- **Icon**: Search icon

## Job progress

- **EstimatedTimeWidget**: Shows expected processing time
- **Live ETA**: Real-time completion estimate during processing
- **Phase tracking**: Progress through analysis phases

## Accepted inputs

**Text formats:**
- Plain text (.txt)
- PDF documents (text extraction)
- CSV with text column
- JSON with text field

**PubMed integration:**
- PubMed IDs (PMIDs) for automatic fetch
- PubMed search queries
- PMC article IDs

## Text preprocessing

- Automatic sentence segmentation
- Tokenization with biomedical vocabulary
- Special character handling
- Unicode normalization

## Model configuration

**Task selection:**
- Named Entity Recognition (NER)
- Relation Extraction (RE)
- Question Answering (QA)

**Entity type filtering:**
- Genes/Proteins
- Diseases
- Chemicals/Drugs
- Species
- Cell types
- Mutations

**Confidence thresholds:**
- Minimum confidence for entity output
- Default: 0.5 (adjustable 0.1-0.9)

## Input limits

- **Text length**: Up to 512 tokens per segment
- **Batch size**: Up to 1000 documents per job
- **File size**: Maximum 50MB per upload

## Read next