FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2907170183> ?p ?o ?g. }

• W2907170183 abstract "Commentary7 January 2019free access The nature of the biological material and the irreproducibility problem in biomedical research George V Papamokos George V Papamokos [email protected] orcid.org/0000-0002-7671-2798 Biomedical Division, The Institute of Molecular Biology and Biotechnology, FORTH-ITE, Heraklion, Crete, Greece Laboratory of Biological Chemistry, Medical School, University of Ioannina, Ioannina, Greece Laboratory of Soft Matter, Department of Physics, University of Ioannina, Ioannina, Greece Department of Physics and School of Engineering and Applied Sciences, Harvard University, Cambridge, MA, USA Search for more papers by this author George V Papamokos George V Papamokos [email protected] orcid.org/0000-0002-7671-2798 Biomedical Division, The Institute of Molecular Biology and Biotechnology, FORTH-ITE, Heraklion, Crete, Greece Laboratory of Biological Chemistry, Medical School, University of Ioannina, Ioannina, Greece Laboratory of Soft Matter, Department of Physics, University of Ioannina, Ioannina, Greece Department of Physics and School of Engineering and Applied Sciences, Harvard University, Cambridge, MA, USA Search for more papers by this author Author Information George V Papamokos1,2,3,4 1Biomedical Division, The Institute of Molecular Biology and Biotechnology, FORTH-ITE, Heraklion, Crete, Greece 2Laboratory of Biological Chemistry, Medical School, University of Ioannina, Ioannina, Greece 3Laboratory of Soft Matter, Department of Physics, University of Ioannina, Ioannina, Greece 4Department of Physics and School of Engineering and Applied Sciences, Harvard University, Cambridge, MA, USA The EMBO Journal (2019)38:e101011https://doi.org/10.15252/embj.2018101011 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Biomedical research has a reproducibility problem since various crucial landmark papers could not be independently reproduced. While there are many causes related to statistical analysis, methodology or insufficient reporting of experimental details, this commentary argues that the complexity of biological material itself is, until now, a largely ignored source of irreproducibility. By discussing examples from evolutionary biology, intrinsically disordered proteins and current biomedical research, it contends that some results are irreproducible because we do not have the knowledge, the tools or the analytical ability to understand biological complexity and how it can give rise to different results. Instead of casting irreproducible research out as bad or sloppy science, they should serve as an inspiration for pioneering research not just to develop such tools but also to attempt to explore what lies beneath our current inability to deal with complexity. Reproducibility is a central dogma of science as expressed by Karl Popper: “… for we have seen that non-reproducible single occurrences are of no significance to science”. However, the life sciences do have an irreproducibility problem for published research articles: during the past years, evidence of irreproducible results has been accumulating, and various methodological causes have been identified (Eisner, 2018) and extensively reviewed (see Eisner, 2018 and references therein). Yet, the role of the unique nature of biological material itself has not been discussed analytically for its potential role in the irreproducibility problem. I suggest that irreproducibility may also be a combined result of the complexity of the biological material and of our finite ability to deal with it, provided that all other factors are addressed (Eisner, 2018). Irreproducibility in carefully selected experiments can also be interpreted to the end that our current methods, theories and techniques are either applied ipsilaterally or are insufficient to analyse and describe functions, properties and unknown or missing parts of the biological material. To support this argument, I will first present the perception of the science of biology by Ernst Mayr, one of the leading evolutionary biologists of the past century who described the evolutionary process as unpredictable or, at best, only statistically predictable (Mayr, 1961). Subsequent recent advancements either challenge the impossible or probabilistic approach of repeatability and predictability at certain levels, or suggest that predictable biochemical changes in evolutionary biology do not have a predictable molecular basis. A second argument is the discovery of a whole new kind of proteins, intrinsically disordered proteins (IDPs) (Dyson & Wright, 2005), that are key factors in health, disease, evolution and now possible drug targets. Lastly, two key examples at the level of eukaryotic cells (Hines William et al, 2014) and at the organismal level (Lithgow et al, 2017) that were unable to account for different states adopted by biological material upon infinitesimal changes in the microenvironment which support that this is an opportunity to produce new science. I will also introduce methodology bias as a new term for the discussion of reproducibility. Perception of biology by Ernst Mayr Ernst Mayr, in his classic work “The Growth of Biological Thought”, epitomized the philosophy of research in biology. He argued that the complexity of biological systems inevitably places biology beyond the limits of the natural sciences as the diversity of biological systems is based on rules that are substantially different from the laws of physics. Mayr believed that these rules cannot be described by methodological reductionism and can only approached probabilistically, and that mathematics, although fundamental in physics, can contribute less to biology. In his landmark paper Cause and Effect in Biology (Mayr, 1961), Mayr described biology as consisting of two different schools of thought: functional biology aims to answer the question “How?” while evolutionary biology attempts to answer “Why?” To this end, he used a warbler on his summer place in New Hampshire as an example to distinguish between proximate causes that are linked to functional biology and ultimate causes linked to evolutionary biology. Thus, the physiological condition of the bird—its sensing of photoperiodicity and temperature—is the proximate cause of migration, while the reduced availability of food during winter and its genetic disposition are the ultimate causes of migration. In the same paper, Mayr concluded that “… causality in biological systems is not predictive or at best is only statistically predictive”. Two years earlier, at the 1959 Cold Spring Harbor Symposium on Quantitative Biology, he had questioned not only the contribution of mathematics to biology, but also argued that applying mathematics in evolution is fundamentally and methodologically wrong: “I am impressed by the uniqueness, by the unpredictability, and by the unrepeatability of evolutionary events. Let me end this discussion with the provocative question: Is it not perhaps a basic error of methodology to apply such a generalizing technique as mathematics to a field of unique events, as is organic evolution?” Eventually, his claim that the “generalizing technique” of mathematics is not applicable to biology and evolution was refuted by other scientists (Crow, 2009). Today, fields such as bioinformatics, computational biology or molecular modelling indeed show that mathematics can be applied to analyse and understand biological evolution; during Mayr's time, it was the human inability to perform highly complex calculations and the lack of sufficiently powerful computers that prevented the application of sophisticated mathematical modelling in biology. By way of example, computational phylogenetics, which is a part of systematics, Mayr's field, demonstrates how human limitations were enhanced by computer science—software and hardware. More generally, computational science enhanced our ability to solve complex problems by orders of magnitude and encouraged scientists to develop new methodologies that none would have conceived before, simply because the technology was not available. Last, but not least, we have entered the era of big data which requires mathematics and bioinformatics to make sense of the huge amount of information. In general, back in the 1950s, as biology was maturing into an autonomous research field, many scientists and philosophers, such as Michael Scriven, accepted that, in evolutionary biology, “satisfactory explanation of the past is possible even when prediction of the future is impossible”. Indeed, applying the laws and theories on complex problems in biology, psychology, anthropology, history, cosmology, economics and quantum physics increasingly required statistical analysis that inevitably yielded errors. For Mayr, unpredictability and the unrepeatability in biology were therefore almost a dogma. Recent findings in molecular biology challenge this dogma as well as the various examples of convergent evolution in biology. Adaptive convergent evolution entails the independent emergence of similar or identical traits in multiple lineages that undergo the same ecological stimulus (Zou & Zhang, 2015), At the molecular level, proteins exhibit convergence too. When different ancestral amino acids are substituted by the same descendant amino acid along independent evolutionary lineages, convergent substitution can be identified. When the same ancestral amino acid is changed to a newer amino acid along independent lineages, a parallel substitution takes place (Zou & Zhang, 2015). Another example is the observation that many insects that feed on Apocynaceae plants evolved independently the ability to avoid the toxic effects of cardenolides, chemicals produced by the same plants. Zhen et al analysed the alpha subunit of the sodium pump, Na+, K+-ATPase (ATPa) which is the protein target for cardenolides in a broad range of taxa and found evidence of parallel changes and duplications that explain the shift to avoid toxicity. Their findings support the hypothesis that adaptation follows evolutionary paths that minimize negative pleiotropy (See Zhen et al, 2012; Zou & Zhang, 2015). This raises an important question in evolutionary biology: Is evolution indeed unpredictable and unrepeatable or are our knowledge, theories and methods still insufficient? There is a fundamental difference though: if unpredictability and unrepeatability are properties of the evolutionary process, then these properties must be systematically verified. If these are unverified properties that appear occasionally and in conflict, we must examine whether unpredictability and unrepeatability fall into our current human inability to predict. This inability might be a temporary and can be revoked in the future. The latter would be a clear sign that more research is needed. So far, recent works support that evolution is unpredictable at the molecular level. For instance, Natarajan et al studied the predictability of genetic adaptation by examining the molecular basis of convergence in haemoglobin function of 56 avian taxa with different altitudinal range limits—more specifically, they tested whether high-altitude taxa have convergently evolved increases in Hb-O2 affinity and whether this evolutionary process is linked to parallel amino acid substitutions (Natarajan et al, 2016). The authors showed that predictable changes in biochemical phenotype do not have a predictable molecular basis, since only a few changes were attributable to parallel amino acid substitutions at key residues, while the majority of them were attributable to nonreplicated substitutions and/or parallel substitutions at sites that are not considered “key residues”. While Mayr's claim about the value of mathematics proved wrong, his observation (Mayr, 1988) about the complexity of biological systems is critical: at the molecular level, biomacromolecules and macromolecules of inanimate material do not differ and show similar physicochemical behaviour—this is evident from various studies especially for simple models. But biomacromolecules possess unique properties. In a living organism, they are hierarchically ordered to form new entities: biomolecules to cells, cells to tissues, tissues to organs and organs to fully functional systems. When a fully functional system is assembled, new unique functions and properties appear that can only be found in living organisms: response to external stimuli, metabolism, growth, differentiation and replication (Mayr, 1988). The last experiment described above (Natarajan et al, 2016), along with Mayr's critical observation, show that adaptive changes may be consequences of various paths at the molecular level (different mutations with different mechanisms), which converge to a common endpoint. They show that the biological material has the necessary complexity to create different processes upon the same stimulus for reasons not yet fully understood. It raises a fascinating question: Is there a way to simulate reversible routes from the endpoint to the initial states of living organisms that evolved? Intrinsically disordered proteins Intrinsically disordered proteins (IDPs) are a class of proteins that do not adopt a defined secondary structure; they possess flat energy landscapes; and their spatial representation resembles a polypeptidic sequence fluctuating between various conformations such as extended coils or collapsed globules. (Dyson & Wright, 2005). Despite their dynamic nature, they are multifunctional and can exhibit various interactions depending on cell-signalling and regulatory networks. Their primary structure is dominated by charged, polar, small hydrophilic amino acids, and they are either entirely disordered or a part of the sequence constitutes an intrinsically disordered region (IDR). IDPs are subject to post-translational modifications, which adds more complexity to IDPs’ functionality. Interestingly, IDPs are highly abundant in nature and their abundance increases with the complexity of the organism from bacteria to archaea to eukaryota: more than half of all eukaryotic proteins are estimated to embody at least regions of intrinsic disorder (IDRs). IDP functions play a central role in many diseases: cancer, neurogenerative diseases, cardiovascular diseases, type II diabetes and acquired immunodeficiency syndrome, while there are numerous findings that they are also part of the evolutionary process (Uversky et al, 2014). For example, the IDP protein α-synuclein is linked to Parkinson's disease, dementia with Lewy bodies, Alzheimer's disease and Down's syndrome. Unfortunately, a lot of information about IDPs was lost or ignored. The standard method to determine the structure of a protein has been X-ray crystallography, but many proteins were not investigated because they failed to crystallize; since these were considered negative results, this information was not published. Moreover, many proteins had parts missing in their crystal structure because these were comprised of intrinsically disordered regions that failed to crystallize. These parts were ignored until recently, which is an example of methodology bias: the use of a method that can give reliable results (X-ray crystallography) provided that the material under study adopts a specific state or condition (well-folded protein). However, if the material cannot adopt this state or condition owing to unknown properties (IDPs, IDRs), it leads to methodology bias and misleading results, wrong analysis and lost information. Another example of methodology bias is irreproducibility of experiments involving enzymes that are caused by differences in purification procedures and assay methods. Moreover, molecules that inhibit a specific enzyme may initiate multiple downstream effects, because they can also inhibit other enzymes the activity of which is not monitored (The last two sentences are corroborating comments of an unknown reviewer). Mayr's view that mathematics cannot be applied to biology can also be interpreted as a methodology bias. The result of this methodology bias is that we are missing a lot of information on IDPs, which has a severe effect on our ability to gain knowledge in living organisms. Immanuel Kant, in his “The Critique of Judgment” written in 1790, described an organism as “… a whole which result from the functioning of the parts, while the parts, in turn, depend on the functioning of the whole”. In fact, the removal of a part may be lethal for the organism and for part. Our lack of knowledge on IDPs is similarly a lack of knowledge on the functionality of an important part that prevents us from understanding and predicting the function of the whole living system. Unexplained irreproducibility in cells and living organisms From the wealth of literature on irreproducibility, I here discuss two papers that highlight and emphasize the sensitivity of the biological material to environmental factors no matter how minute these are (Bissell, 2013). The first example describes how cells from the same human breast cell line but from different sources respond differently to the same assay. Second, studying the role of glucose uptake in cancer progression, Bissell and collaborators showed that either changes in media glucose levels or the cells’ shape, when media was kept constant, changed the nature of the metabolites and the metabolic pathways. A third example from a collaborator of hers showed that they could not reproduce their own experiments using nonmalignant human breast cell lines obtained from an investigator with cells obtained from a cell bank. Subsequent analysis revealed that the cultured cells had drifted and revealed crucial information in cell cycle regulation of the drifted cells. The same researcher and her collaborators provided another excellent example of how to track down and eliminate causes of irreproducibility: working side by side on the same tumour biopsy, they found that even small differences in cell isolation—vigorous stirring versus prolonged gentle rocking—resulted in irreproducible research (Hines William et al, 2014). Gordon J. Lithgow was one of the authors of a 2000 paper describing that a drug-like molecule could extend an animal's lifespan (Lithgow et al, 2017 and refs 1,2 therein). This finding could not be reproduced by other laboratories, and the reason is still unknown (Lithgow et al, 2017). The authors spent 4 years and studied more than 100,000 worms to systematically test ageing interventions in the nematode Caenorhabditis elegans and eliminated many causes of variability. Yet, even when performing identical experiments in a single laboratory, they observed that some cohorts of worms could fall into one of two modes of ageing: short-lived or long-lived. The reason is not understood, and the team is focusing on molecular differences that may account for this difference within the same strain. They state that this phenomenon would have been undiscoverable if they had not eliminated all other sources of variability. In both examples, irreproducibility results spurred deeper analysis at the molecular level that proved or may prove very fruitful. Both examples also show that biological material from cells to living organisms has the property to exhibit different states at almost identical conditions. What do cases A, B and C share? These cases highlight the complexity of the biological material and a critical consequence: biological material can either produce the same result following various molecular routes, or the same stimulus may have different results. How can this consequence affect science and irreproducible research? Suppose that a living system, examined at the molecular level, can exhibit various states in health, disease or evolution: S1, S2 and S3. Starting from either of these states, it can converge to a state denoted as S4. Suppose now that three independent laboratories, starting from state S4, explore the reasons for this reaction and reveal states S1, S2 and S3, respectively. Although these results may appear as irreproducible, they manifest a chance for further research, new knowledge and the missing factor of biological complexity. The current inability of our methods and approaches to deal with the complexity of the biological material and its functions should indeed inspire scientific progress to replace earlier views based on ignorance with solid knowledge based on new findings, methods and approaches. Failures to reproduce experiments should therefore not be considered a catastrophe for science and purged from the literature. Science has made enormous progress through failures and errors: for example, the Rayleigh–Jeans law, the classical approach to describe black-body radiation, resulted in the ultraviolet catastrophe. As the law predicted that cool objects should radiate in the visible and ultraviolet regions of the electromagnetic spectrum, darkness could not exist. Max Planck eventually brought darkness back with the introduction of the quantization of energy. Envoy Many methods and tools that are applied to analyse biological material were originally designed to study materials of the inanimate world. The biological material, however, has unique properties that are not predictable from its constituting parts. Are these methods and techniques sufficient to reveal the secrets of life in health and disease? Do selected irreproducible results imply the need to improve existing and to invent new methods, techniques, theories and approaches? This commentary presents evidence and argues that we need to look out of our current scientific toolbox. This must not be perceived as an excuse for irreproducible research. On the contrary, it shows that when dealing with complex material, we need to share data; describe our work in detail; improve our methods, practices and protocols; increase our collaboration and our interdisciplinary research; and deepen our perception of the field. Questionable practices on a material we only know it partially, on a material that can produce multiple results by default, do not serve science; they just serve career goals. Instead, they demonstrate how sloppy or biased research incubates irreproducible research and builds expensive castles in the sand. This kind of research is fundamentally different from the research and its results I described in this article. We must distinguish between bad science and honest failures that indicate the need for further research. Conflict of interest The authors declare that they have no conflict of interest. References Bissell M (2013) Reproducibility: the risks of the replication drive. Nature 503: 333–334CrossrefPubMedWeb of Science®Google Scholar Crow JF (2009) Mayr, mathematics and the study of evolution. J Biol 8: 13CrossrefPubMedGoogle Scholar Dyson HJ, Wright PE (2005) Intrinsically unstructured proteins and their functions. Nat Rev Mol Cell Biol 6: 197–208CrossrefCASPubMedWeb of Science®Google Scholar Eisner DA (2018) Reproducibility of science: fraud, impact factors and carelessness. J Mol Cell Cardiol 114: 364–368CrossrefCASPubMedWeb of Science®Google Scholar Hines William C, Su Y, Kuhn I, Polyak K, Bissell Mina J (2014) Sorting out the FACS: a devil in the details. Cell Rep 6: 779–781CrossrefCASPubMedWeb of Science®Google Scholar Lithgow G, Driscoll M, Phillips P (2017) A long journey toward reproducible results. Nature 548: 387–388CrossrefCASPubMedWeb of Science®Google Scholar Mayr E (1961) Cause and effect in biology. Science 134: 1501–1506CrossrefCASPubMedWeb of Science®Google Scholar Mayr E (1988) Toward a new philosophy of biology: observations of an evolutionist. Cambridge, MA: Harvard University PressGoogle Scholar Natarajan C, Hoffmann FG, Weber RE, Fago A, Witt CC, Storz JF (2016) Predictable convergence in hemoglobin function has unpredictable molecular underpinnings. Science 354: 336–339CrossrefCASPubMedWeb of Science®Google Scholar Uversky VN, Dave V, Iakoucheva LM, Malaney P, Metallo SJ, Pathak RR, Joerger AC (2014) Pathological unfoldomics of uncontrolled chaos: intrinsically disordered proteins and human diseases. Chem Rev 114: 6844–6879CrossrefCASPubMedWeb of Science®Google Scholar Zhen Y, Aardema ML, Medina EM, Schumer M, Andolfatto P (2012) Parallel Molecular Evolution in an Herbivore Community. Science 337: 1634–1637CrossrefCASPubMedWeb of Science®Google Scholar Zou Z, Zhang J (2015) Are convergent and parallel amino acid substitutions in protein evolution more prevalent than neutral expectations? Mol Biol Evol 32: 2085–2096CrossrefCASPubMedWeb of Science®Google Scholar Previous ArticleNext Article Read MoreAbout the coverClose modalView large imageVolume 38,Issue 4,15 February 2019Caption: EM micrograph of human airway organoid showing cellular organization and distribution of ciliated, secretory and basal cells. By Norman Sachs, Hans Clevers and colleagues: Long‐term expanding human airway organoids for disease modelling. Scientific image by Nino Iakobachvili, Maastricht University. Volume 38Issue 415 February 2019In this issue ReferencesRelatedDetailsLoading ..." @default.
• W2907170183 created "2019-01-11" @default.
• W2907170183 creator A5035915461 @default.
• W2907170183 date "2019-01-07" @default.
• W2907170183 modified "2023-10-13" @default.
• W2907170183 title "The nature of the biological material and the irreproducibility problem in biomedical research" @default.
• W2907170183 cites W2003816086 @default.
• W2907170183 cites W2006192061 @default.
• W2907170183 cites W2021579219 @default.
• W2907170183 cites W2047116672 @default.
• W2907170183 cites W2048785675 @default.
• W2907170183 cites W2070865170 @default.
• W2907170183 cites W2086189298 @default.
• W2907170183 cites W2122527741 @default.
• W2907170183 cites W2143499154 @default.
• W2907170183 cites W2538451863 @default.
• W2907170183 cites W2747311934 @default.
• W2907170183 cites W2765493561 @default.
• W2907170183 doi "https://doi.org/10.15252/embj.2018101011" @default.
• W2907170183 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/6376269" @default.
• W2907170183 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/30617085" @default.
• W2907170183 hasPublicationYear "2019" @default.
• W2907170183 type Work @default.
• W2907170183 sameAs 2907170183 @default.
• W2907170183 citedByCount "6" @default.
• W2907170183 countsByYear W29071701832021 @default.
• W2907170183 countsByYear W29071701832022 @default.
• W2907170183 countsByYear W29071701832023 @default.
• W2907170183 crossrefType "journal-article" @default.
• W2907170183 hasAuthorship W2907170183A5035915461 @default.
• W2907170183 hasBestOaLocation W29071701831 @default.
• W2907170183 hasConcept C127413603 @default.
• W2907170183 hasConcept C55587333 @default.
• W2907170183 hasConcept C70721500 @default.
• W2907170183 hasConcept C86803240 @default.
• W2907170183 hasConceptScore W2907170183C127413603 @default.
• W2907170183 hasConceptScore W2907170183C55587333 @default.
• W2907170183 hasConceptScore W2907170183C70721500 @default.
• W2907170183 hasConceptScore W2907170183C86803240 @default.
• W2907170183 hasIssue "4" @default.
• W2907170183 hasLocation W29071701831 @default.
• W2907170183 hasLocation W29071701832 @default.
• W2907170183 hasLocation W29071701833 @default.
• W2907170183 hasLocation W29071701834 @default.
• W2907170183 hasOpenAccess W2907170183 @default.
• W2907170183 hasPrimaryLocation W29071701831 @default.
• W2907170183 hasRelatedWork W1641042124 @default.
• W2907170183 hasRelatedWork W1990804418 @default.
• W2907170183 hasRelatedWork W1993764875 @default.
• W2907170183 hasRelatedWork W2013243191 @default.
• W2907170183 hasRelatedWork W2046158694 @default.
• W2907170183 hasRelatedWork W2082860237 @default.
• W2907170183 hasRelatedWork W2117258802 @default.
• W2907170183 hasRelatedWork W2130076355 @default.
• W2907170183 hasRelatedWork W2151865869 @default.
• W2907170183 hasRelatedWork W4234157524 @default.
• W2907170183 hasVolume "38" @default.
• W2907170183 isParatext "false" @default.
• W2907170183 isRetracted "false" @default.
• W2907170183 magId "2907170183" @default.
• W2907170183 workType "article" @default.