---
title: Oncology Analytics quickstart
description: >-
  Step-by-step workflows for TMB/MSI calculation, MHC binding prediction, and
  neoantigen prediction.
tags:
  - oncology
  - quickstart
---

> **Retired tools**

> `hrd_score`, `liquid_biopsy_ctdna` and `tumor_fraction_estimate` are retired and can no longer be run. Use an allele-specific copy-number HRD caller such as scarHRD for HRD, ichorCNA on low-pass WGS for tumor fraction, and a dedicated ctDNA/MRD pipeline for ctDNA.

You want to characterize a tumor and predict immunotherapy biomarkers; these workflows walk you from somatic variant or peptide data to clinically anchored readouts like TMB/MSI and neoantigens, each in a few steps. See [Oncology Analytics on the platform](/docs/oncology) for tool details.

## Oncology Analytics quickstart

This guide covers three common workflows: calculating TMB/MSI, predicting MHC binding, and identifying neoantigens.

## Workflow 1: Calculate TMB and MSI status

### Step 1: Provide somatic mutations

```
Calculate the TMB and MSI status from these somatic variants
```

Or configure manually:

| Parameter | Value |
|-----------|-------|
| Input type | VCF or mutation list |
| Genome build | GRCh38 |
| Panel size (Mb) | 1.5 (for targeted panels) |

### Step 2: Review TMB and MSI results

| Field | Example Value |
|-------|---------------|
| TMB | 14.2 mut/Mb |
| TMB Status | High |
| Total Mutations | 213 |
| Coding Mutations | 186 |
| MSI Score | 32.4% |
| MSI Status | MSI-H |
| Unstable Microsatellites | 18 / 56 |

### Step 3: Interpret the results

| Biomarker | Threshold | Interpretation |
|-----------|-----------|----------------|
| TMB ≥ 10 mut/Mb | TMB-High | Checkpoint inhibitor eligible |
| TMB < 10 mut/Mb | TMB-Low | Standard of care |
| MSI-H (>= 30% unstable) | Microsatellite instable | Pembrolizumab eligible |
| MSI-L (10-30% unstable) | Low instability | Not MSI-H eligible |
| MSS (< 10% unstable) | Microsatellite stable | Not MSI eligible |

## Workflow 2: Predict MHC binding

### Step 1: Provide peptides and HLA types

```
Predict MHC binding for these peptides with HLA-A*02:01
```

Or configure:

| Parameter | Value |
|-----------|-------|
| Peptides | List of 8-11mer sequences |
| HLA alleles | `["HLA-A*02:01", "HLA-B*07:02"]` |
| MHC class | I |

### Step 2: Review binding predictions

| Peptide | HLA Allele | IC50 (nM) | Percentile Rank | Classification |
|---------|-----------|-----------|-----------------|----------------|
| YLQPRTFLL | HLA-A*02:01 | 12.4 | 0.08 | Strong binder |
| KLPDDFTGCV | HLA-A*02:01 | 234.5 | 1.2 | Weak binder |
| SPRWYFYYL | HLA-B*07:02 | 8.1 | 0.04 | Strong binder |

### Step 3: Interpret binding affinity

| IC50 Range | Percentile Rank | Classification |
|------------|----------------|----------------|
| &lt; 50 nM | &lt; 0.5% | Strong binder |
| 50-500 nM | 0.5-2% | Weak binder |
| &gt; 500 nM | &gt; 2% | Non-binder |

## Workflow 3: Predict neoantigens

### Step 1: Provide somatic mutations and HLA types

```
Identify neoantigen candidates from these somatic mutations for HLA-A*02:01
```

Or configure:

| Parameter | Value |
|-----------|-------|
| Mutations | VCF or mutation list |
| HLA alleles | `["HLA-A*02:01", "HLA-A*03:01", "HLA-B*07:02"]` |
| Expression data | RNA-seq TPM values (optional) |
| Min binding affinity | 500 nM |

### Step 2: Review neoantigen candidates

| Rank | Gene | Mutation | Neoantigen Peptide | HLA Allele | IC50 (nM) | Expression (TPM) |
|------|------|----------|-------------------|-----------|-----------|------------------|
| 1 | KRAS | G12D | KLVVVGADGV | HLA-A*02:01 | 18.3 | 42.1 |
| 2 | TP53 | R175H | HMTEVVRHC | HLA-A*02:01 | 45.2 | 28.7 |
| 3 | BRAF | V600E | KIGDFGLATV | HLA-A*02:01 | 67.8 | 15.3 |

### Step 3: Prioritize candidates

Candidates are ranked by a composite score incorporating MHC binding affinity, expression level, clonality, and mutation type. Top candidates with strong binding (IC50 &lt; 50 nM) and high expression are prioritized for vaccine design.

## Next steps

- See [inputs](/docs/oncology/inputs) for all parameters, file formats, and HLA type specifications
- See [outputs](/docs/oncology/outputs) for result field descriptions and threshold interpretation
- Use [Genomics](/docs/genomics) for upstream variant annotation and gene information
- Use the Compound Passport for therapy selection based on biomarker results

## Read next