A single E. coli cell weighs about a picogram. That number matters more than it looks

On 10 July 2026, at 19:35 UTC, the biophysicist Niko McCarty posted a deceptively simple observation to X: a single Escherichia coli bacterium weighs about one picogram, roughly equal to the mass of the DNA packed inside a single hummingbird cell. The comparison, made inside a thread about how widely cell mass varies even within a single species, lit up timelines across the biology community for what it implies about measurement rather than about microbes.

The point is not that bacteria are heavy or hummingbirds are light. The point is that "how much does a cell weigh?" is a question with a six-figure answer depending on which cell, in which organism, at which point in the cell cycle. That uncertainty is the bottleneck for an entire generation of diagnostics, drug discovery and synthetic-biology tools that promise to read disease or build tissues one cell at a time.

A number that moves

McCarty's thread, which circulated under the handle @nikomccarty, opens with the E. coli baseline: about one picogram per cell, the canonical figure cited in cell-biology textbooks. It then walks through why that figure misleads. A human liver cell can weigh roughly four to five nanograms, around four thousand times the mass of an E. coli. A fat cell tips the scales higher. A fertilised egg is heavier still, and an unfertilised one is lighter. The DNA content of a single hummingbird cell, McCarty notes, comes in at roughly the same order of magnitude as the entire wet mass of a bacterium, a coincidence that exposes how much of "cell weight" is actually nucleic acid, water and protein scaffolding rather than anything resembling a uniform substance.

For working biologists the discomfort is familiar. Cells are not interchangeable widgets. They swell during division, contract after, load up on lipid droplets or empty out depending on metabolic state. A measurement taken at one moment in a population may not describe the same cell an hour later, even within a clonal line. The field has learned to live with averages, but the tools now arriving on the bench do not need averages. They need single-cell resolution, in grams.

Why the picogram is suddenly a bottleneck

Three threads of work are converging on cell mass as a measurable variable rather than an inferred one. Quantitative phase imaging, which infers mass from light delay through a transparent cell, has matured to the point where instruments can return a mass estimate on every cell in a microscope field of view in real time. Suspended microchannel resonators weigh single cells one at a time as they pass through a vibrating cantilever, with attogram-level sensitivity. And a newer class of microfluidic "mass cytometers" sorts cells by weight before downstream analysis, the way fluorescence-activated sorters sort by colour.

The diagnostic pitch is straightforward: many diseases change cell mass before they change cell shape. Cancer cells in early-stage tumours tend to be larger and more variable than their healthy neighbours. Infected cells swell. Stem cells preparing to divide pack on protein. A mass signature, read label-free and at speed, could in principle flag trouble before a stain or a genetic test would.

The catch is calibration. Quantitative phase imaging returns mass as an estimate that depends on the assumed refractive index of the cell's interior. That assumption drifts with protein content. Suspended microchannel resonators are exquisitely precise relative to themselves but need a reference bead, and the beads drift. Every published dataset on cell mass is, somewhere, sitting on top of a calibration choice that the field has not standardised.

The counter-read

Not everyone in the cell-biology community treats single-cell mass as the next big biomarker. Sceptics argue that mass is a downstream proxy for the things researchers actually care about: protein content, DNA ploidy, mitochondrial activity, membrane potential. Mass can be derived, roughly, from those variables; if the underlying signals are measurable, mass adds little. There is also a practical objection: instruments that weigh single cells one at a time do not scale to clinical throughput, and imaging-based mass estimators produce so much data per experiment that downstream analysis becomes its own bottleneck.

The strongest defence of the approach, articulated in work from labs at the Rowland Institute, ETH Zurich and several US systems-biology groups over the past decade, is that mass is the only quantity one can read non-destructively, in real time, across an entire living population. Genetic tests require lysis. Stains perturb the system. Mass measurement can sit on a microscope stage and watch cells decide, in real time, whether to divide, differentiate or die. That longitudinal record, the argument runs, is what gives the picogram its analytical bite.

What hangs on the resolution

If mass-based diagnostics mature, the early applications will almost certainly be in oncology and reproductive medicine, where the existing single-cell assays are expensive and slow. A biopsy read by phase imaging could in principle return a mass distribution and a malignancy flag in the time it takes a pathologist to fix and stain the slide. Synthetic-biology labs, which today characterise engineered cell lines by fluorescence and growth rate, would gain a cheap, label-free quality-control readout.

The structural question underneath is whether the field will adopt a shared mass standard. McCarty's thread functions as a quiet appeal for exactly that. Until instruments, beads and refractive-index assumptions are calibrated to a common reference, every lab's "nanogram per cell" is a private measurement. A community-agreed standard would do for single-cell mass what standardised fluorescence channels did for flow cytometry in the 1980s: turn a comparative art into a quantitative science.

The picture that emerges from the thread is not a story about a bacterium. It is a story about a measurement, and the slow, unglamorous work of making it good enough to trust.

Wire provenance

This editorial synthesis draws on the following public wire/social posts:

• https://x.com/nikomccarty/status/1944034987239530767
• https://en.wikipedia.org/wiki/Quantitative_phase_imaging
• https://en.wikipedia.org/wiki/Suspended_microchannel_resonator
• https://en.wikipedia.org/wiki/E._coli