As Howard Says: Remembering Howard Shapiro

By Jeff Carrell - September 21, 2022

19-220880-LS_Journals_Critical Values_Flow Cytometry_CV

The international flow cytometry community was saddened by the passing of Howard Shapiro on November 10, 2021. Many of his colleagues, acquaintances, and admirers went online and shared their personal stories, acknowledged his massive contribution to our field, and fondly recalled memorable encounters.

Most felt sadness for the loss of a pioneer and proselytizer, but the unanimous appreciation of Howard’s wit was striking. He used poems and songs (often accompanying himself on guitar) to explain cytometry, and his lectures were must-see events at local cytometry society meetings and the annual CYTO meeting. He became a one-name icon; it is not uncommon to overhear “Howard says…” at a cytometry gathering, and everyone knows that Howard always means Howard Shapiro.

His classic text Practical Cytometry (Wiley, ISBN: 978-0-471-41125-3, now in its 4th edition) has become an oft-quoted source and required reading for cytometrists around the world. Among the charms of the book are his Laws of Cytometry, which oddly enough are scattered throughout the book non-chronologically. We can all learn from these truisms, whether we use flow cytometry in clinical settings, academic research, public health, or industrial biopharma. Reviewing the laws is my humble memorial.

First: A 51μm particle clogs a 50 μm orifice!

Synopsis: This is largely self-evident; clogs impede the flow in flow cytometry. Remember cytometers rely on laminar flow or hydrodynamic focusing to align the cells or particles of interest as they pass the lasers. Disturbances to laminar flow can result in the core stream deviating from the central position of the flow cell and different velocities within the core; turbulent flow results in chaos in your data! Most importantly, a clogged cytometer means down time, repairs, lost experiments, and annoyed colleagues.

Second: (Flip Wilson’s Rule): What You See is What You Get!

Synopsis: This law relates to the ability of an optical system to resolve particles you want to observe (your passing cells) from the background in which they are moving (sheath fluid). If the laser beam is illuminating a volume of the sheath stream that is many-fold larger than the particles within, the cells must be brighter than the fluid to be resolved from background. More importantly, this law points out the importance of knowing about sources of noise, whether they are electronic or biological. If you cannot resolve signal from noise, WYSIWYG: just noise.

Third: What’s in the Bottle Isn’t Necessarily What’s on the Label

Synopsis: While reagents for flow cytometry have improved over the last five decades, in the long history of cytometry there have been many instances of ‘Lies, damn lies, and biological stains.’ This can be extended to modern application: we should not assume the new lot of fluor-conjugated antibody will perform like the lot we’ve been using for the last few years. We can cite recent recall reagent recalls: conjugations can vary between lots, tandem dyes can break down, and microbe contamination can foul reactivity. Consider ‘the bottle’ when troubleshooting.

Fourth: (The Supermarket Theorem): Most Babies Aren’t Born in Supermarkets!

Synopsis: This one is tougher conceptually. When you are out shopping, you may see women with an infant or two, or toddler(s), or some combination of kids, and maybe you might see a few pregnant women. But rarely (or never) have you seen a woman at the Food Lion in the midst of going from childless to having a child, so it follows that the process happens elsewhere. When we observe distinct antigenic clusters on cells (for example cells in blood are either CD3+ or CD3- without a continuum of intermediate CD3) that indicates that the differentiation occurs somewhere else. Cells do not go from CD3 negative to very positive for CD3 in an instant jump. Thus, the transition either happens elsewhere, or it is very rare where we are looking. If you do see a continuum of antigenic expression (CD45RO comes to mind) then it suggests the process of changing is occurring where we are looking.

Fifth: (The Barber Shop Theorem): No Man Walks into a Barbershop With a Long Beard Who Hasn’t had a Shorter One, and No Barber Can Make a Beard a Lot Shorter Without Cutting It!

Synopsis: As with the Fourth Law, this refers to distributions of measured cell properties, and how we analyze them. We don’t go from hairless to bearded in one big leap, and measured cell properties don’t change in leaps either. Further, one can trim hair a little or a lot or shave entirely; unsupervised algorithms can find clusters but dividing a continuum may be a matter of taste. As Howard says, ‘surface antigens aren’t nature’s version of a bar code.’ This is something to consider as we move to automated analysis.

Sixth: There are Some Cell Identification Problems That Even Monoclonal Antibodies Can’t Solve!

Synopsis: This is really an extension of the Fifth Law and is a vestige of simpler times when flow analysis was limited to fewer fluors and simple two-dimensional gate hierarchies to define ‘populations’ of leukocytes. It is true that we can add valuable supplemental information to flow analysis with nucleotide probes, membrane dyes, metabolic-reactive stains or other fluorescent reagents to identify structures or processes, and now we have an ever-growing palette of analytical tools that take analysis well beyond 2D space.

Seventh: No Data Analysis Technique Can Make Good Data Out of Bad Data!

Synopsis: Howard points out that ‘a great many people spend a lot of time with sophisticated numerical methods’ in order to violate this law. We must know the limits of our system, and more complex does not mean better; data analysis should be guided by some knowledge of the biology under study, and a strong notion of what the experimenter wants to know. Before performing an analysis, the goals should be known, and the methods should be matched to the goals.

Eighth: Know Thy Cells!

Synopsis: Again, match the analysis tools to the type of analysis needed. This applies to all aspects of cytometry including instrument selection. ‘If you are trying to sort very big or very small stuff, be sure the machine can do it; if you are looking for very faint signals, try before you buy.’ Always consider the questions you need to answer when choosing a system, an experimental plan, or an analysis. 

Zeroth: There Is No Magic.

Walt Disney has been quoted as saying “There is no magic in magic; it’s all in the details.” Howard Shapiro’s most universal law simply states the fundamental nature of science and scientists: Despite our awe of Nature’s complexity, when we ask the right questions and rigorously test hypotheses with our ever-expanding toolbox, the trickiest of systems may be revealed. We are all privileged to view the world of cytometry from Howard’s shoulders. 

Jeff Carrell

Manager, CCR-Frederick Flow Core