Notes about writing and reading treatises about biology#

First version: 2020-03-23
Last update: 2022-03-02
Persistent link to latest version:

1 Foundational approach

A discipline can be prestened and studied bottom-up or top-down. The bottom-up approach starts by laying the foundational principles, then expresses complex results in terms of simplest constituents. With a pure bottom-up approach, there is a small base of knowledge which is taken for granted (in mathematics, these are the axioms and definitions). Every other concept is introduced in terms of previously presented concepts. This approach is best illustrated by mathematics and to a lesser degree physics. For example: in physics, classical mechanics is the foundation that enable the acceleration of a rocket to be calculated from its mass, flow rate of the propellent and exhaust speed of the propellent.

In biology the foundational approach is of limited application because for many systems of interest it is not fully known how their behavior arises from its parts or from its interaction with other systems. For example: It is known that most of the effects of estrogens are mediated by the nuclear receptors ERα and ERβ. It is known that estradiol (the main endogenous estrogen) binds to ERα and ERβ and that this modifies genetic expression. It is not know with full detail which genes are affected nor how the altered genetic expression gives rise to all the perceptible effects of endogenous or exogenous estrogens. Thus it is not possible to give an exposition of the physiology of sex hormones in a fully foundational bottom-up approach. It is often the case that medication enters widespread use before all its pharmacological targets are known. In our example of estrogens: Exogenous estrogens including naturaly occurring (estradiol, estrone, estriol) and with human-made structures (diethylstilbestrol, ethynylestradiol) were already in widespread use as anticonceptives when the ERβ receptor was discovered [1].

2 About references

The purposes of providing references in a treatise are:

  • To enable the reader to verify the claims, ideally following the sequence of referenced works until the primary source.
  • To direct the reader for further information about the topic.

Unless it is very clear otherwise, explicitly mention the relation of a reference to the topic in prose instead of parenthetically. Present the reference in a conscise way that answers the following questions:

  • Does it provide support for a claim made in the present article or further reading?
  • If it supports a claim, which claim?
  • If it is for further reading, what is it about?

To merely specify the reference parenthetically often leaves unclear what is the relevance of the reference; thus the reader has to waste time checking whether it is of any relevance in his/her case.

For claims which do not depend on subjective judgement, cite the primary source. Example: Suppose one wants to state the biological half-life for a compound. If one finds it in a review and this review cites a primary source which evaluated the pharmacokinetic parameters of the compound, then cite the primary source, not the review. The primary source contains a detailed description of the experiment and the results with the least distortion; therefore it is more relevant in relation to the half-life. If the review is useful, then cite it separately (e.g.: “[primary source] found the half-life of [compound] is [...]. [reference] reviewed the pharmacology of [compound]”).

When available, include quantitative data. E.g.: “[reference] found that norfluoxetine has a half-life of 14 days”, not “[reference] found that norfluoxetine has a very long half-life”. The former is a stronger statement from which the latter follows.

3 Note-taking

When researching a topic within biology that is not thoroughly understood and the scientific knowledge is scattered in many scientific articles, take notes of the relevant articles and their main findings in a coherent document. Consider publishing this document so that it may be useful to other interested parties. Taking notes is not neccessary for topics for which there are comprehensive treatises.

Example 1: The anatomy and histology of the liver is thoroughly known and described in books like Tortora, Derrickson (2017).

Example 2: Knowledge about treatments that reduce pigmentation of skin, hair and eyes is scattered among hundreds of papers. We took notes and published them as a review that provides a complete or almost complete outline of these treatments and their mechanisms of actions.

4 Epistemological soundness

The goal of science is to make predictions about objectively observable properties of the real world.

4.1 Vagueness

Vague concepts can be useful as heuristical tools to help interested parties in arriving at scientific conclusions and in thier application; they are not scientific knowledge and must not be presented as such.

For example, “irritable bowel syndrome” is a vague non-scientic concept. It is useful as a heading under which related articles are grouped and can be found more easily. Any experiment done on irritable bowel syndrome should define a concrete criterion to be used in its stead. This is also called operationalization. The term “operationalization” is misleading because it suggest the error that “irratable bowel syndrome” is the primary entity of scientific study and the concrete definitions are secondary [2].

4.2 Illusory differences without empirical distinction

Assertions that appear to speak about the real world and do not make any objectively verifiable prediction are not scientific hypothesis. Example: In physics, special relativity and aether theory with Lorentz transform make exactly the same predictions about the outcome of measurements in any experiment. It is outside of science to argue “which one is true”. Note that the Lorentz transform used in relativity was originally derived in the context of aether theory.

5 Distinguish evidence-based claims from authority-based claims

Scientific rigor requires that claims are supported by evidence. An almost-universal mistake in papers about biology is that claims that are supported by evidence in a scientifically rigorous way are mixed without distinction with claims made on the author’s subjective interpretation. To distinguish them, find the primary source for the claim, examine the objective observations reported and ignore the author’s interpretation. The objective observations are nearly always numerical and often reported in tables or graphs.

When the popular media reports about new scientific findings it often reports only the author’s interpretation and ignores the objective results. This is filtering the science out of scientific practice.

Some authority-based claims are considered “common knowledge” in medical practice. Examples:

  • That monoamine oxidase inhibitors interact with almost literally every medication and food in life-threatheing ways.
  • That humans “need” 8 hours of sleep. There is the additional mistake of vagueness. What objective observations are predicted if this “need” is not met?

6 Avoid mixing subjective judgements with scientific content

Avoid mixing subjective judgments of the authors with the scientific content. Do not start an article with a non-technical justification of the importance of the treatise. There are 2 main reasons for this:

  • The reader is looking for information on a given topic (as opposed to opening books or articles at random). The reader already has his/her own justification to be interested in this topic.
  • For justification that depends on subjective values, values are not science and have only a marginal place in scientific treatises. They are tolerated, rather than required. Note that this applies to the common values in the discipline at the time of writing the article as much as it applies to the values particular to the authors of the treatise.

For example, articles about volitional use of cross-sex hormones often start by justifying the intervention in terms of treating a supposed mental disorder. This conflates the science and technique of endocrinology with the pseudoscience of psychiatry. Unfortunately, the malpractice of conflating subjective judgement with scientifically established facts is pervasive in medicine. This can make the problem not apparent to workers limited to this area. By contrast, consider an hypothetical article that presents a more efficient system for transmission of electric power that starts by giving a justification in terms of addressing class struggle. To the electrical engineer it is obvious that the talk of class struggle in this hypothetical paper is a digression and out of place. Yet today’s medic fails to see that in the same way the talk of supposed mental disorders and self-identities is likewise out of place.

7 Units and physical quantities

7.1 Units

The International System of Units (SI for its acronym in French) is the humanity-wide standard rational system of units suitable for practical scientific purposes. Treatises must use SI units or units allowed for use within the SI. The choice for SI units as opposed to British imperial or US customary units is not arbitrary. Among them, only the SI has the property that all units within the system that are of the same dimension differ by a factor that is a power of 10; therefore, they can be converted by shifting the decimal point or simple addition to the exponent in scientific notation.

Blood pressure. Expressing blood pressure only in millimetres of mercury is unacceptable. Express blood pressure using the pascal or its multiples.

Heart rate. The minute is an unit accepted for use within the SI; therefore, it is acceptable to express heart frequency in units of inverse minutes (symbol: 1/min or min−1).

Nutritional energy. Expressing nutritional energy in calories is unacceptable. There are several units called “calorie” that differ by over a factor of 1 000. Use the joule (symbol: J) or its multiples.

The symbol for microgram is “μg”. Both letters are upright, not italic nor cursive. Any other symbol or abbreviation is incorrect and inacceptable. It is widespread mistake to refer to the unit microgram as “mcg” or “microg”.

7.2 Salts

When writing the amount of a pharmaceutical that can form salts, write which salt. If it is the free pharmaceutical (not a salt), write that. Otherwise there is uncertainty of the dose because an unknown fraction of the mass is the counterion. For example: caffeine is used as free and as caffeine citrate.

Caffeine formMolecular massRelative potency
Caffeine (free/anhydrous)194.19 g/mol1 (by definition)
Caffeine citrate386.31 g/mol0.503

It is standard to assume that the counterion is itself biologically inert except in studies that examine the difference in effect among different counterions for the same free pharmaceutical.

Assuming that the counterion is biologically inert, to compute the relative potency among different salts or the free form, use the formula:

  • Potency of B relative to A = Molecular mass of A / Molecular mass of B


  • Potency of caffeine citrate relative to free caffeine
    = (194.19 g/mol) / (386.31 g/mol)
    = 0.503

The conversion then proceeds as if it were converison of units. Example:

  • Give the equivalent of 200 mg of caffeine citrate as free caffeine.
    Answer: 200 mg of caffeine citrate are equivalent to 200 mg · 0.503 = 101 mg of caffeine citrate.
  • Give the equivalent of 400 mg of free caffeine as caffeine citrate.
    Answer: 400 mg of caffeine citrate are equivalent to 400 mg / 0.503 = 795 mg of caffeine citrate.

8 Generalization

Biology studies very complex systems whose behavior depends on conditions that can not be fully fixed yet do not vary with a well-characterized random distribution in the relevant scenario. Much of the behavior of biological systems that humanity is interested in can not be characterized with a measurement. For example: severity of an illness; only in a few cases there exists an objective marker (e.g.: blood concentration creatinine for kidney dysfunction, blood concentration of clotting factors for hemophilia).

In practice one has to use knowledge gained under specific experimental conditions to make choices under different conditions. One has to generalize, that is, one assumes that the system in question will behave in a similar way under similar conditions conditions without having experimentally confirmed that this is the case. Typically generalization works. The importance of being aware that one is depending on generalizations to be able to identify when things does not work because the generalization one used does not work.

8.1 Inter-individual genetic variation

Genetic variation contributes uncertainty that is rarely accounted for in the literature. For example: Garza-Flores (2014) compared the pharmacokinetics of esters of estradiol administered as depot injections; this study used 2 medical centers, one in Latin America and another in Asia. This study found different pharmacokinetic parameters between the 2 study centers. It is very likely this is in part caused by genetic variation between the populations.

An especially salient example of genetic variation breaking generalization is the taste of phenylthiourea: some people experience it as bitter and others as tasteless. In a review, Risso et al. (2016) found that this variation is mostly attributable to a single gene: TAS2R38.

The notion that women have 46,XX genotype and men have 46,XY genotype is an oversimplification. Counterexamples include:

  • Male phenotype with natural 46,XX genotype. E.g.: Ryan et al (2013).
  • Female phenotype with 46,XY genotype beacuse of a loss of function allele in the nuclear androgen receptor.
  • Genotypes other than 46,XY and 46,XX.
  • Transsexuals change their phenotype to the other sex with the use of cross-sex hormones and/or surgery.

8.2 Inter-individual anatomical variation

There exists anatomical variations that cause no functional difference and therefore go unnoticed for most individuals. These differences become relevant when performing surgery. For example, the fact that some people have no frontal sinus is relevant to facial feminization surgery for MTF transsexuals. Deschamps-Braly (2018) states that in people without frontal sinus, forehead reduction can be accomplished by burring whereas people with frontal sinus usually require forehead reconstruction.

8.3 Inter-species variation

For many reasons, species other than humans are used to research the effect of treatments intended to be used in humans. An example of where this fails follows: Tyrosinase is the main enzyme involved in the generation of the pigmentation of skin, hair and eyes. Tyrosinase from mushrooms has been used to screen compounds for inhibitors of tyrosinase. Mann et al. (2018) found that the molecular structure required to strongly inhibit mushroom tyrosinase is different from that required to strongly inhibit human tyrosinase. They found that in humans thiamidol is a much stronger inhibitor of tyrosinase that kojic acid; in mushrooms it is the opposite.

9 Thresholds for the definition of a disease

The boundary between healthy and pathological is ultimately subjective. When there are objective markers for a pathological biological condition there is subjective judgement involved in chosing a threshold for the definition of the disease. For some pathological conditions there are no “in-between” cases in practice; all cases are clearly healthy or clearly pathological. For example: bone fractures. In practice a bone is either intact or broken all or most of the way (greenstick fractures). Some other conditions have in-between cases in practice. For example: hyperthyroidism. The threshold between pathological hyperthyroidism and sub-clinical hyperthyroidism is a matter of subjective judgement.

10 Fischerian hypothesis testing

p-values are not the probability that the hypothesis is false given the observation; popular media often make this mistake in presenting scientific findings. p-values are the probability that the observation is false given the hypothesis is also false. As-is, this does not tell what is the probability that the hypothesis is true.

Why is the probability that the hypothesis is true not routinely given in scientific papers? Becuase computing it requires the use of Bayesian statistics. Bayesian statistics require as input a detailed model of prior beliefs about the possible outcomes of the experiment. In simple cases there exists a model which is obviously neutral. In most cases found in practice, chosing a model among several plausible ones is a matter of judgement. In these cases, chosing among prior probabilities is a matter of judgement. Thus we do not give a recommendation to avoid fischerian hypothesis testing, only to be aware of its counterintuitive meaning.

11 Caveats about terminology

11.1 Adrenaline (“epinephrine”)

Use “adrenaline” instead of “epinephrine”. Both names reference the kidneys. Only the name “adrenaline” is consistent with the use of “renal”/“adrenal” to refer to the kidneys.

11.2 “alternative medicine”

The term “alternative medicine” is ill-defined. In common usage it referes to a class of techniques which purport to treat disease which have in common that they are based on magical thinking or contradict basic laws of physics. This includes homeopathy, crystal healing, use of static magnetic fields. The term is unfortunate because it focuses on the fact that these purported treatments are an alternative from mainstream medicine, which is irrelevant by itself. The relevant property that they have in common is that they can be immediately inferred to not to work because they are incompatible with elementary principles of the laws of the natural world.

11.2.1 Plants

Use of plants with the intention to treat disease is often included under the category of alternative medicine; this carries the erroneous implication that consumption of plants has no pharmacological effect. Some plants contain pharmacologically active ingredients in meaningful amounts. A well-known example is caffeine in tea and coffee which has a stimulating effect.

Plants have been used with the intention to treat disease since before the advent of modern medicine. Effectiveness can not be confirmed nor ruled out in the general case. The effectiveness of a plant species to elicit an effect should be evaluated on a case per case basis with the same scientific methods expected to evaluate the effectiveness novel synthetic pharmaceuticals.

11.3 “associated”

Instead of “X is associated with Y” or “X has been implicated in Y” write “X and Y are correlated”, “X causes Y”, “Y causes X” or “X and Y have a common cause” as it corresponds.

The term “associated” is very common in the biology literature. For example: “[X lifestyle choice] is associated with [Y disease]”. The term “implicated” is equivalent. This usage is misleading because it invites a conflation between a correlation and a causal relationship. Statements like this are commonly used to express a correlation found through observational studies which are insufficient to show a causal relationship.

We suspect that the reason for the poplularity of this phrasing is intellectual cheating: It gives the impression of presenting strong knowledge (a causal relationship) while paying the burden of proving a much weaker statement (a correlation).

11.4 “heavy metals”

Instead of “heavy metals” write the concrete elements and the numerical amount if known. Example: Write “We found that FooBar brand bread contains a mass fraction 2.4×10−5 of lead”, not “We found that FooBar brand bread contains heavy metals”.

Exactly which metals are “heavy metals”? There is no standard definition. The term is nearly meaningless. Is iron a heavy metal? The term does not say. It is used to elicit an emotional reaction of being toxic. It has no place in a scientific context.

11.5 Irreversible inhibitors

Irreversible inhibitors do not have a permanent effect. The “irreversible” refers to the mechanism of action as molecular level. Irreversible inhibitors create a covalent bound with the active site of the enzyme, effectively clogging it. At an organism level the effect is reversible because the body renews enzymes in the span of weeks.

11.6 Mistaking definitions with causal relationship

Avoid writing that an observation is caused by a condition that is defined to entail the observation.

Example: A local anesthetic is defined as a substance that suppresses pain sensation where it is applied. The sentence “topical lidocaine relieves pain because it is a local anesthetic” is equivalent to “lidocaine relieves pain where it is applied because it relieves pain where it is applied” which is a tautology and therefore conveys no information about the real world. Instead write the real cause: “Lidocaine relieves pain because it blocks voltage-gated sodium channels and therefore suppresses the propagation of action potentials that signal pain”.

11.7 Psychic

In the context of study of humans, “psychic” is synonymous with “psychological”. The term “psychic” was more common in the first 3 quarters of the 20th century for this usage. It should not be confused with the everyday meaning of “psychic” as purported supernatural powers.

11.8 Race

Do not hesitate to explicitly refer to race in a scientific context when it is relevant to the discussion.

Because of an undue influence of political correctness, workers are sometimes reluctant to use the word “race” and often misuse “ethnicity” instead. Race is genetic; ethnicity is a combination of genetics and culture. An attempt at a rational explaination for this avoidance is the claim that races have no scientific basis. This is a deception. It is well established that many traits that are part of the classical conception of race (facial proportions, stature, pigmentation of skin, hair and eyes) are genetically determined and cluster by broad geographic region after accounting for recent migrations. The precise number and boundaries of races are subjective. That is not a reason to discard the concept. In biology many concepts have subjective boundaries including many well-accepted diseases and syndromes. E.g: hyperthyroidism, irritable bowel syndrome. The singling of race among them is hypocritical and purely because of political reasons that have no place in science.

11.9 Seropositive, seronegative, seroconversion

“seropositive” is a person that has antibodies for an antigen mentioned in the context. “seronegative” is somebody that does not have antibodies for that antigen. “seroconversion” is the event where the immune system starts to produce those antibodies.

Laypeople often erroneously believe that “seropositive” means “infected with HIV”. The term “seropositive” is not specific to HIV. It is possible to develop antibodies for a pathogen without being infected with it. The principle of vaccines is to cause the immune system to generate antibodies to counter a disease without causing the disease. It is possible to have an infection without having antibodies against the pathogen. Many pathogens produce an infection before the immune system produces antibodies against them. When an infection is cured it is usually the case that the antibodies against the pathogen that causes it persist for years or life-long. Antibodies are not always against a pathogen. In autoimmune diseases the antigen is a normal part of the body.

11.10 “significant”

To state that an effect is statistically significant, always write “statistically significant” in full. There are 2 meanings to “significant”:

  • The effect is strong enough to be relevant. Opposite of an effect being negligible.
  • Statistically significant. When using Fischerian statistics, an observed effect is said to be “significant” when the probability of the effect exceeding this threshold given the null hypothesis is smaller than an arbitrarily chosen probability, usually 0.05.

11.11 “There is no evidence”

Avoid writing “there is no evidence that X”.

If what is meant is that X is false, then write that and provide a reasoning. Example: Instead of “There is no evidence that crystal healing has any effect other than placebo.”. write “From basic physics it follows that crystal healing has no effect other than placebo.”.

If a search was made for evidence and none was found then write that. Example: Instead of “There is no evidence that drinking tea daily reduces risk of autoimmune diseases.” write ”We searched PubMed using the query “tea autoimmune” and we did not find any experiment that confirms this nor any proposed biological mechanism”.

12 Footnotes

  1. The discovery of ERβ was presented in Kuiper et al. (1996). See Koehler et al. (2005) for a review that mentions the previous paper.
  2. .
  3. A similar consideration applies to capturing intuitive concepts in mathematical definitions. The thesis that the mathematical definition captures the intuitive concept is not strictly speaking mathematics becuase it concerns a mathematical entity and an intuitive non-mathematical entity. Instead that thesis is pre-mathematical. See Szabó (2018) for an account of the comments of Kalmár László regarding the mathematical definition of effectively computable functions which include this consideration.
  4. .

13 References

  • J. C. Deschamps-Braly (2018) “Facial Gender Confirmation Surgery Facial Feminization Surgery and Facial Masculinization Surgery”. DOI: 10.1016/j.cps.2018.03.005.
  • J. Garza-Flores (1994) “Pharmacokinetics of once-a-month injectable contraceptives”. DOI: 10.1016/0010-7824(94)90032-9.
  • K. F. Koehler et al. (2005) “Reflections on the Discovery and Significance of Estrogen Receptor β”. DOI: 10.1210/er.2004-0027.
  • G. G. J. M. Kuiper et al. (1996) “Cloning of a novel estrogen receptor expressed in rat prostate and ovary”. DOI: 10.1073/pnas.93.12.5925.
  • T. Mann et al. (2018) “Inhibition of Human Tyrosinase Requires Molecular Motifs Distinctively Different from Mushroom Tyrosinase”. DOI: 10.1016/j.jid.2018.01.019. Open access.
  • D. S. Risso et al. “Global diversity in the TAS2R38 bitter taste receptor: revisiting a classic evolutionary PROPosal”. DOI: 10.1038/srep25506. Open access.
  • N. A. Ryan et al. (2013) “A case report of an incidental finding of a 46,XX, SRY-negative male with masculine phenotype during standard fertility workup with review of the literature and proposed immediate and long-term management guidance”. DOI: 10.1016/j.fertnstert.2012.11.040.
  • M. Szabó (2018) “Kalmár’s Argument Against the Plausibility of Church’s Thesis”. DOI: 10.1080/01445340.2017.1396520. Full text in ResearchGate.
  • G. J. Tortora, B. Derrickson (2017) “Principles of anatomy & phisiology”. ISBN (electronic version): 978-1-119-32064-7.