Opens in a new tab

Chemoproteomics for covalent drug discovery

Find what your compound hits, measure how much of your target it engages, and confirm the answer with a second, independent LC-MS method, all in one lab.

~20,000
reactive cysteine sites quantified per experiment
>99%
target engagement still resolved
3
complementary methods, run on one platform

What it answers

What is my compound hitting?

Global cysteine profiling measures competitive reactivity across about 20,000 reactive cysteine sites per experiment. It reveals selectivity and finds the targets of phenotypic hits.

How much of my target is engaged?

Targeted assays with heavy internal standards quantify engagement precisely, and still resolve it above 99%.

Is the signal real?

Every cysteine measurement can be paired with a protein-level measurement from the same sample, separating true engagement from a protein that is simply changing in abundance.

Methods

Three ways to measure covalent engagement

Each method answers a different question, and each suits a different point in a program.

Discovery · unbiased

Global cysteine profiling

What it measures

Competitive labeling of cysteines across the whole proteome, read out by TMT.

Best for

Screening libraries, mapping selectivity, and finding the target of a phenotypically active compound.

  • Rapid, unbiased selectivity
  • Turnkey, with no assay development
  • Harder on low-abundance targets
  • Engagement is inferred
Targeted · indirect

Loss of the unmodified peptide

What it measures

The unmodified target peptide disappearing, tracked by a PRM panel with heavy standards against stable control peptides.

Best for

Routine, sensitive engagement readouts, including in clinically relevant matrices.

  • Sensitive and robust
  • Works in clinically relevant matrices
  • Requires assay development
  • Engagement is inferred
Targeted · direct

Adduct quantification

What it measures

The compound–peptide adduct itself, as a fraction of total target peptide signal.

Best for

Confirming engagement directly, including near-complete engagement.

  • Engagement measured, not inferred
  • Resolves near-complete engagement
  • The most demanding to develop
  • Needs active, properly folded protein

Why run more than one

One system, three methods, one answer

Many providers offer one of these approaches. We run all three on the same system, so each result can be checked against the others.

In our sotorasib work, the directly measured adduct fraction matched the engagement inferred from loss of the unmodified peptide. Once two independent methods agree, the faster indirect assay can carry the rest of a program with confidence.

Direct and indirect, cross-checked: two readouts, one answer.

Case study

Sotorasib in H358 cells

We ran all three methods on sotorasib, the KRAS G12C inhibitor, in H358 cells. Baltier et al., AACR 2025.

01 · Global cysteine profiling

Found the target, and ruled out a false hit

Six doses in triplicate fit in one TMT 18‑plex, quantifying more than 8,000 cysteine peptides. The KRAS G12C cysteine peptide disappeared with dose while the KRAS protein held flat, so the loss reflects engagement, not degradation.

A cysteine peptide on vimentin (C328) also dropped with dose and looked like an off-target hit. The companion proteome showed the vimentin protein itself was changing, which ruled it out.

>8,000
cysteine peptides
6 × 3
doses × replicates, one 18‑plex
Bar charts of KRAS G12C peptide falling and KRAS protein flat across six sotorasib doses in triplicate.Bar charts of VIME C328 peptide and VIME protein falling together, and the protein-normalized peptide staying flat.
Global cysteine profiling of sotorasib in H358 cells. Top: the KRAS G12C peptide falls with dose while KRAS protein stays flat. Bottom: VIME C328 appears to fall, but VIME protein falls in step; normalizing to protein removes the apparent hit.

02 · Targeted engagement

Measured engagement two ways, and they agreed

The directly measured adduct fraction tracked the engagement inferred from loss of the unmodified peptide. Engagement above 99% was still resolved, across a 4-log linear range.

Isoforms are the hard part. Peptides shared across RAS isoforms moved with dose, while the KRAS-unique peptide SFEDIHHYR held steady. Normalizing to that unique peptide is what makes the engagement number trustworthy.

>99%
engagement resolved
4 logs
linear range, R² 0.99
Stacked bar chart of adducted versus unmodified G12C peptide across seven sotorasib doses, the adducted share rising to about 97%.
Measured directly, the sotorasib-adducted G12C peptide rises to about 97% of total G12C signal by 0.5 µM.

How it works

How global cysteine profiling works

Everything runs in 96-well format, with on-deck lysis and automated liquid handling, in our Framingham laboratory.

  1. 1

    Treat

    Cells treated across your dose series. Cell culture available in house.

  2. 2

    Lyse natively

    Proteins extracted under native conditions, with Covaris on-deck lysis. We recommend sending cell pellets; lysates are also accepted.

  3. 3

    Label cysteines

    Cysteines left free by your compound capped with desthiobiotin-iodoacetamide.

  4. 4

    Digest & barcode

    Each sample digested and TMT-labeled, then pooled into one 18‑plex.

  5. 5

    Our edge

    Enrich & split

    Two streams from one sample:

    Captured: cysteine peptides
    Flow‑through: companion proteome
  6. 6

    Acquire & analyze

    LC-MS on our Orbitrap platforms, then quantification, QC and a report.

Specifications

What to expect

Specification
Global cysteine profiling
Targeted engagement assays
Input per sample
Global cysteine profiling
500 µg protein
Targeted engagement assays
10 µg protein
Multiplexing
Global cysteine profiling
TMT 18‑plex, for example 6 doses in triplicate
Targeted engagement assays
Not applicable
Capacity
Global cysteine profiling
Up to 4 × 18‑plex, 72 samples
Targeted engagement assays
Up to 200 samples
Turnaround
Global cysteine profiling
About 4 weeks from project start
Targeted engagement assays
About 3 weeks; new assays about 10 weeks to develop
Assay development
Global cysteine profiling
None needed
Targeted engagement assays
Once per target, then reused across the program

Every project includes

  • Quantitative tables as Excel and flat files
  • QC data, including labeling efficiency and normalization factors
  • Methods and summary report; methods text for publication on request
  • RAW mass spectrometry files on request
  • A consultation on what the data shows

Evidence

Published work

Our method · AACR 2025

An LC-MS toolbox for identifying and guiding the development of cysteine-targeted small molecule covalent inhibitors

Baltier, Erickson, Kunz, Braun and Erickson. AACR Annual Meeting 2025, Chicago. See the poster

Customer program · Science 2025

BBO-10203 inhibits tumor growth without inducing hyperglycemia by blocking RAS-PI3Kα interaction

Simanshu DK, et al. Read more

Customer program · Nature 2019

The clinical KRAS(G12C) inhibitor AMG 510 drives anti-tumour immunity

Canon J, et al. Read more

Thirty peer-reviewed papers draw on data from our laboratory. See all publications, or read our technical note, Covalent inhibitor profiling.

Working with us

From first conversation to delivered dataset

  1. 01

    Consult

    We help design the dose series, plexing scheme, replicates and controls before any sample is committed.

  2. 02

    Submit

    Ship Monday to Wednesday, overnight on dry ice, with the submission form.

  3. 03

    Execute

    Preparation, enrichment, acquisition and analysis all run in our Framingham laboratory. Nothing is subcontracted.

  4. 04

    Deliver

    Processed tables, a methods and summary report, and a consultation on what the data shows.

Planning a covalent program?

Tell us about your compound, target and samples. We’ll recommend the right method, or combination, before any sample is committed.