Is Science Driving Technology?
An Investigation in the Field of Breast Cancer:
Analysis of Polymerase Chain Reaction and CRISPR/Cas9

doi:10.21203/rs.3.rs-5348962/v1

Download PDF

Research Article

Is Science Driving Technology? An Investigation in the Field of Breast Cancer: Analysis of Polymerase Chain Reaction and CRISPR/Cas9

https://doi.org/10.21203/rs.3.rs-5348962/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

The debate about whether science drives technology or conversely remains unresolved. Various approaches have been proposed to determine the driving force and influence of science on technology. Numerous technologies have been analyzed and compared, and in some cases, science has been shown to drive technological progress, particularly in the field of biochemistry. However, whether the influence of science on different technologies within the same sector changes over time has not yet been investigated. This study analyzes how the influence of science on technologies in the field of breast cancer therapy has evolved over the years, using polymerase chain reaction and CRISPR/Cas9 as examples.

For the study, two data sets were created for each technology: one for scientific publications and the other for patents. This approach allowed for the inclusion of approximately 1,800 articles and 400 patents.

It was observed that in the case of CRISPR, the scientific literature is only slightly ahead of the patents. In addition, a shift in the patenting strategy of China and the USA was observed: from predominantly scientific activity to increased patenting.

Although these two technologies are in a science-intensive field, there is little to no evidence of a significant time lag. The country analysis revealed that China, in particular, is increasingly focusing on commercial applications, more so than was previously observed with PCR.

Approximately two decades ago the Human Genome Project achieved a significant accomplishment: the publication of the complete sequencing of human DNA. This pivotal event represents a significant advancement in genetics and biomedical research, providing insights into our biological identity. Moreover, it serves as an example of the efficacy of international collaboration, demonstrating how collective endeavors can facilitate the advancement of groundbreaking research(Hood & Rowen, 2013; Watson, 1990).

The comprehensive sequencing of the human genome inaugurated a new era of scientific inquiry, catalyzing significant progress in medical science and the development of innovative technologies. Nevertheless, despite these advances, the age-old question remains: which is the driving force behind progress, science or technology? The resolution of this query remains a formidable challenge, prompting the exploration of diverse methodologies, including the renowned analysis of unpatented references.

Previously methodologies for investigating the interplay between science and technology were developed. Numerous studies employed patent citation analysis to examine the contribution of science to technological development (Gittelman & Kogut, 2003; Narin et al., 1997; Narin & Noma, 1985; Schmoch, 1997; Tijssen, 2001; Van Looy et al., 2003; Van Vianen et al., 1990).

Recent approaches combine analysis of cited publications and patents, using the latter as an indicator of technology-intensive fields. This approach involves dividing the total number of patents in each technology by the number of publications citing them. A similar methodology is employed in the analysis of publications citing patents. In this study the methodology proposed by Willoughby (2021) is employed, utilizing two distinct data sets for patents and scientific articles. This approach allows migration for the study of knowledge without the influence of the respective authors of patents or publications.

The polymerase chain reaction (PCR), developed by Kary Mullis in 1983, has become a standard procedure in molecular biology. It is used to amplify DNA in vitro, making it an essential tool in genomic research, the diagnosis of infectious diseases, and forensic analysis (KB; Mullis, 1994). The clustered regularly interspaced short palindromic repeats (CRISPR) system represents a revolutionary advancement in genetic engineering, first documented by Jennifer Doudna and Emmanuelle Charpentier in 2012. The CRISPR method allows for precise genome editing by cutting DNA at specific locations, enabling researchers to modify genes with unprecedented accuracy. This technology has numerous applications, including genomic research, agriculture, and medicine (Jinek et al., 2012). These technologies together illustrate the intricate relationship between scientific invention and technological innovation, which drive each other forward in an ever-evolving landscape of knowledge and applications.

2.1 Science: Data source and data selection of scientific publications

The publications had to fulfill several criteria for the data to be included in the study. To increase robustness and reliability publications had to be peer-reviewed, written in English, and scientifically relevant. The inclusion criteria for the articles included reaching a threshold of at least 40 citations, which serves as an indicator of scientific relevance (Narin, 1994). The Database was built by using Scopus from Elsevier. To narrow the field down to a suitable Dataset the terms “Polymerase chain reaction” OR “PCR” were used. The included articles also had to include “breast cancer” and “treatment” OR “therapy”. The time horizon was set from 1990 until 2023. The analysis included information such as the year of publication, digital object number, journal, keywords and citations. To ensure uniformity impact factors were consistently standardized to the year 2022. The definition of the impact factor was adopted by Clarivate, which was developed by Eugene Garfield. A scientific journal's impact factor is calculated by dividing the number of citations of articles in this journal in the current year by the total number of articles published in the two previous years (D'Armiento et al., 2019; Garfield, 1955).

The selection process is illustrated in Fig. 1. For CRISPR/Cas9, the first operator was replaced by "CRISPR" OR "CRISPR/Cas9". The other operators remained unchanged. Initially, 1431 publications were considered during the first stage of the selection process using predefined operators. Following a rigorous screening 246 publications ultimately satisfied the predetermined selection criteria.

In the first step, the search operators were selected, and 7,363 publications could be included. In the second step, only journal articles and peer-reviewed articles were considered, and 4685 articles were excluded.

Citations are included in this article for several reasons. On the one hand, a highly cited publication reflects the reputation of the article in the scientific and technical community, thus increasing its trustworthiness. For some authors, the bibliometric method is superior to peer review (Abramo & D'Angelo, 2011). A significant number of citations also often indicates basic research and may indicate emerging trends in research.

2.2 Patents data selection

Data collection was carried out using Minesoft. To ensure comparability the same Boolean search parameters were used as described above. One dataset was thus created for PCR technology and one for CRISPR technology. Patents with the status "granted" and "pending" were included. The granting jurisdiction including the country of the patent was identified for the patents. The two datasets follow distinct chronological timelines. The timeframe for PCR spans from 1993 to 2023, whereas the period for CRISPR extends from 2014 to 2023.

For better comparability, the data were standardized using min-max normalization. The statistical analysis was conducted using RStudio. The normal distribution was examined using Kolmogorov-Smirnov tests and Shapiro-Wilk tests and the correlation using the Pearson and Spearman correlation. The mean values were analyzed by applying the Student's T-test and the Wilcoxon-Mann-Whitney test.

Table 1 summarizes the datasets for PCR and CRISPR patents and publications in both fields. Patents related to PCR technology from 1993 to 2023 were included, while patents related to CRISPR were considered from 2014 to 2023. The journal articles were divided into subgroups according to citations per article. All published patents of the respective technologies were included.

Table 1: Journal Publications and Patents on PCR (1993-2023) and CRISPR (2014-2023).

The table presents data collected for two technologies, including journal publications with over 10 and over 40 citations, as well as the entire data set of granted patents (Mdn: Mean, SD: Standard deviation).

Table 2 presents the statistical analysis, which shows, that the results deviate from a normal distribution. Conversely, the correlation analysis shows a moderate positive correlation between the results. For PCR moderate correlations of 0.52 and 0.57 indicate a medium to strong positive relationship between publications and patents, suggesting that they tend to move in similar directions but are not perfectly synchronized. In contrast CRISPR exhibits only a weak relationship with positive correlations of 0.18 and 0.23, respectively. The mean value comparison clearly shows significant differences between the categories. The significant p-value indicates that the difference between the patents and publications groups in the areas of PCR and CRISPR is not due to a random effect.

3.1 Results for Polymerase Chain Reaction

The dataset contains 4498 publications with more than 10 citations, with a maximum of 326 articles published in one year. However, there is a high fluctuation with a standard deviation of 116.5, the median value was 112 publications. 41% of the articles had a higher citation rate of more than 40. The maximum number of publications fell by more than half to 128 publications, as well as the median value and the standard deviation (Table 1). From Table 3 it can be deduced, that the median impact factor is 5.32 and fluctuates with a standard deviation of 11.4.

The statistics for the 434 patents are less volatile, with a maximum of 31 patents in 2021, a median of 14 and a standard deviation of 8.3.

The cross-correlation shows how patents and publications change over time. The maximum correlation is a lag of 0, which indicates that there is no time lag between the publication of a scientific article and a patent. The same applies to the median and the mean of 0. However, the high standard deviation of 18.33 indicates that the time lag between patents and publications fluctuates greatly (Figure 3).

Table 4 reveals that China has the most journal publications, whereas the number of patents in China ranks behind the USA and India. Compared to other countries, the USA has the most patents and the second-highest number of publications.

3.2 Results for CRISPR/Cas9

The second part of summarizes the data collected for CRISPR Technology. The number of publications with more than 10 citations differs from the PCR technology in some important respects. The period considered is significantly shorter, at 9 years only approximately 30% of the time span of PCR technology. A total of 651 publications were included, with a maximum of 150 publications per year. The standard deviation was 89.2 and the median was 119. At a higher citation limit of 40 citations, the standard deviation decreases by more than 75% compared to 61% for PCR.

A total of 246 fulfilled the criterion, the maximum number of publications per year amounted to 50 publications. The standard deviation and the median are 20.8 and 24 respectively, which also shows a distinct contrast to the PCR technology. In the category of published patents, 56 patents with CRISPR technology were identified. In addition to the temporal distortion, this also reflects the fact that there are 87.15 fewer patents in the CRISPR field. The maximum number of patents per year was 12, the standard deviation was 3.8 and the median was 4.5.

In the computation of cross-correlation the maximum of the coefficient is -1 years, which indicates, that the publications are 1 years prior to the patent publications.

The mean time lag is 0, proving a heterogeneous distribution, that both patents run ahead of publications and vice versa. The standard deviation of 5,63 years demonstrates the large scatter. The median time lag supports the findings of the mean, which is also 0 and proves, that the mean delay is neither strongly influenced nor affected (Figure 4).

The analysis of publications and patents by country activity demonstrates that China and the USA are also comparably active in CRISPR technology. The ratio between the number of publications and patents has changed considerably for the technologies. In the case of PCR, only 17 patents were registered in China out of a total of 435 publications, while the ratio for CRISPR rose from 4% to 22%, i.e. 24 patents were registered for 109 publications. In contrast, in the USA, 206 patents were granted in the field of PCR, and scientific activity accounted for 165 publications. The ratio of patents to publications is 10% for CRISPR, but 126% for PCR (Table 4).

Figure A shows the chronological development of patent publications and scientific articles in the field of polymerase chain reaction (PCR) from 1990 to 2023, with only granted patents shown in green and scientific articles in blue. Figure B shows the chronological development of patent publications and scientific articles in the field of breast cancer for the same period. In this case, patents granted are also shown in green and scientific articles in blue. The data has been normalized for better comparability.

3.3 Nonpatent references

Figure 4 and Table 5 summarize the results of 32 active US patents with PCR technology.

If the chronological horizon of the cited publications (publication date of the article) is analyzed, a time span of an average of 23.4 (Md. = 22) years and an average of 21.7 (Md. = 19) years before the publication of the patent and 2.1 years (Md. = 2) after the publication of the patent can be observed. Figure 4 illustrates this relationship.

The patents under consideration reference an average of 66.1 (Md. = 31) publications.

Comparing the publications cited by the patents with those from the data set of publications, it is noticeable, that only one publication is also cited in the patents. If the publications cited by the patents are compared with each other, it becomes apparent, that the majority of the articles are only cited once and only 22 scientific articles are cited more than three times by the patents analyzed (Table 7).

The cited publications within the patents exhibit a discernible concentration on specific journals as evidenced by the dataset. In terms of journals, the data reveal an average impact factor of 26.3 (Md. = 25.1), with citations per publication averaging 1485.5 (Md. = 876.7). A total of 1525 publications from the 32 active patents were cited from 463 journals.

The results suggest that PCR has played a stable and significant role in breast cancer research over a long period of time, while CRISPR/Cas9 has experienced a rapid and intense upswing. Each technique has its unique advantages: while PCR is known for its simplicity and efficiency in DNA amplification, CRISPR/Cas9 offers unrivaled precision in genome editing.

Comparing patent and publication activities across different technologies, a noticeable trend emerges. Specifically, patent applications within the field of Polymerase Chain Reaction (PCR) and corresponding publications are observed to be declining rapidly. Conversely, in the case of Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR), both patent activity and the number of publications are experiencing a substantial increase.

Several factors may account for these divergent trends. Changes in governmental funding allocations or shifts in research focus prompted by the emergence of new and more promising opportunities could be contributing factors. Further investigations need to be carried out to determine the reason for the change in technologies and possible indicators.

4.1 Polymerase Chain Reaction

The data show no time lag for PCR, but a dynamic correlation between patents and publications over time. The highest correlation at a lag of 0 years indicates a simultaneous relationship between publications and technological advancements. A similar simultaneous or inverse relationship has also been observed in ICT-related technologies. However, Dernis (2016) demonstrated in a cross-technology analysis, that technological developments typically precede publications by approximately 2–3 years. In a study of patents from 1980 to 2000 in the related field of recombinant DNA, Lo (2010) found no detectable time lag, noting, that the average age of publications cited by these patents was 9.8 years.

The lack of delay can be interpreted as an indicator of the close connection between research and development. PCR is an established technology, that is indispensable in many areas of research, serving as a crucial tool for scientific investigations. Due to the advanced maturity of the technology, its availability within research infrastructure is very high, enabling universal application and rapid implementation. Additionally, the close integration of basic research with industrial and commercial applications contributes to the near-simultaneous occurrence of publications and patent filings. Other potential reasons for the absence of a time lag include the standardized nature of PCR protocols and the widespread familiarity with the technology among researchers, which facilitate faster transitions from discovery to practical application.

4.2 Clustered Regularly Interspaced Short Palindromic Repeats

The results of the study indicate that the publication of research findings precedes the patenting of technologies involving CRISPR. However, this effect is not as pronounced as in studies of other technologies (Smirnov & Willoughby, 2021). The comparison of patents and publications shows that both patent activity and publication activity are correlated. This implies that both the number of patents and the number of publications increase or decrease over time, although not in perfect parallel. The cross-correlation indicates a symmetry between the two objects of investigation, namely that publications tend to be higher than patent applications.

4.3 Country Analysis

The analysis reveals that China, and the USA are at the forefront of advancements in PCR and CRISPR technologies. The USA is particularly distinguished by its extensive patent filings and robust commercial applications in the field of PCR. Conversely, in the realm of CRISPR technology, China demonstrates not only substantial scientific engagement but also an increasing number of patent filings, suggesting a strategic emphasis on intellectual property acquisition. This trend aligns with Martin-Laffon's 2019 findings, which documented significant patent activity in CRISPR technology by both, China and the USA.(Martin-Laffon et al., 2019)

Additionally, Europe, although not a leading player, contributes both academically and commercially to the field of PCR. Japan and South Korea also exhibit noteworthy patent activity in CRISPR, with 5 and 7 patents respectively, highlighting a growing commitment to intellectual property protection in this emerging field.

For nanotechnology Shapira and Wang were able to show, that China and the USA receive the highest level of funding in this field worldwide via state institutes such as the National Natural Science Foundation of China or the US National Science Foundation. This high level of funding is also reflected in the large number of publications. (Shapira & Wang, 2010) Lo was also able to show, that the USA received the most patents granted and made the greatest scientific contribution.

The reason for this predominant position of the USA is a cluster of universities and companies. Lo was able to prove, that 181 patents were published by the University of California (Berkley) and 121 scientific articles by Harvard employees. (Lo, 2010)

In the area of the technology examined here, further research should be carried out in the future, firstly to investigate a possible change in the patent and publication strategy. The structures of subsidies should also be investigated in both technologies in order to better understand the steering effect of subsidies.

4.4 Non-Patent References

As part of the analysis of non-patent references we were able to prove, that the polymerase chain reaction and CRISPR technologies are among the most science-intensive areas of biotechnology. This fits into the scientific landscape. Meyer and Van Looy et al. were already able to show, that technologies in biotechnology are patent science intensive. (Callaert et al., 2006; Meyer, 2000; Narin & Noma, 1985; Van Looy et al., 2003)

This study demonstrates that the patents reference a highly diverse range of publications, with very few publications being cited more than once. It also reveals that the dataset of the literature is almost independent of the publications cited in the patents. Van Raan's 2017 study corroborates this finding, showing that only 3–4% of the articles were indexed by Web of Science or Scopus. In addition to the number of articles it can be seen, that articles cited in patents concentrate on certain journals, so there was a focus on a few journals. (Van Raan, 2017; Van Vianen et al., 1990).

Considering the publication dates of the cited articles there is an average gap of 4 years (Md.: 3 years) between the publication of the patent and most of the cited articles. This indicates a close relationship with scientific research and aligns with the time lag analysis, which showed no detectable time lag between publications and patents. These findings are consistent with the study by Narin et al.(Narin et al., 1995)

When examining the number of citations per individual article, a high citation count per article is evident. This indicates both the scientific significance and the foundational nature of the research, which has been cited by patents. Collins & Wyatt and Kwon et al. demonstrated that citation counts are high in rapidly growing trends within scientifically intensive research fields. This also applies to PCR, which is retrospectively considered an emerging technology. (Collins & Wyatt, 1988; Kwon et al., 2019)

In our study we verified, that US patents are strongly biased towards American journals: 8 out of 10 cited journals were from the US, while only two were from the UK. These journals publish 27% of the cited articles, a national bias that Narin et al. demonstrated in 1997. (Narin et al., 1997) However, it should be noted, that the most important journals with very high impact factors are from the US.

Compared to Lo's study, there is a 60% concordance for the 10 most frequently cited articles. This high level of correlation supports the hypothesis, that the most frequently cited journals predominantly publish fundamental research.

Further research is needed to investigate the citation mechanisms and strategies in patents. The question arises as to whether these journals are cited because they publish the research required for the patent or because scientists publish in these journals due to their reputation. The former in particular is a reasonable effect, as basic research can have a significantly greater reach as it can be applied in many scientific fields.

The analysis highlights the distinctive trajectories and roles of PCR and CRISPR technologies in scientific research and patent activity. Polymerase chain reaction (PCR) has retained a significant and stable presence in breast cancer research due to its simplicity and efficiency in DNA amplification. In contrast, CRISPR/Cas9 has experienced a rapid surge in popularity, driven by its unparalleled precision in genome editing.

The study reveals divergent trends in patent and publication activities. With regard to PCR, both patents and publications are in decline, reflecting the established status of the technology and the possibility of saturation in the field. Conversely, CRISPR shows a substantial increase in both patents and publications, indicative of its emerging and dynamic nature.

A detailed analysis indicates that there is no significant time lag between the patenting of PCR and the publication of related research. This suggests a close integration of research and application. In contrast, the publication of CRISPR-related articles typically precedes the filing of patents by approximately two years. This suggests that the transition from research findings to patentable technologies occurs relatively rapidly.

A country-specific analysis reveals that the USA and China are the leading players in both technologies. The USA excel in PCR patents and commercialization, while China demonstrates a strong scientific and increasing patent activity in CRISPR. Notably, Europe, Japan, and South Korea also contribute significantly, particularly in the field of CRISPR patents.

Future research should focus on understanding the shifting strategies in patenting and publication, as well as the impact of funding structures on technological advancements. The study also emphasizes the significance of prompt scientific dissemination in stimulating innovation and safeguarding intellectual property, particularly in rapidly evolving fields such as CRISPR.

For the non-patent references in 32 active US patents, it can be shown, that the patents for PCR technologies cover a long-time horizon, on average 23.4 years before the patent. This is contrary to other studies in the field of biotechnology.(Huang et al., 2015)

It also shows, that the patents cite many scientific references, on average 66.1 (Mdn.: 31).(Gao & Guan, 2009; Huang et al., 2015) It also shows that patents in the field of biotechnology are cited from a few journals (M.: 39.2; Mdn.: 23) with a high impact factor. For a more precise analysis of the relationship, a network analysis should be carried out in the future.

Abramo G, D'Angelo CA (2011) Evaluating research: from informed peer review to bibliometrics. Scientometrics 87(3):499–514
Callaert J, Van Looy B, Verbeek A, Debackere K, Thijs B (2006) Traces of prior art: An analysis of non-patent references found in patent documents. Scientometrics 69(1):3–20
Collins P, Wyatt S (1988) Citations in patents to the basic research literature. Res Policy 17(2):65–74
D'Armiento J, Witte SS, Dutt K, Wall M, McAllister G (2019) Achieving women's equity in academic medicine: challenging the standards. Lancet 393(10171):e15–e16. https://doi.org/10.1016/S0140-6736(19)30234-X. & Columbia University Senate Commission on the Status of
Gao X, Guan J (2009) Networks of scientific journals: An exploration of Chinese patent data. Scientometrics 80(1):283–302
Garfield E (1955) Citation indexes for science: A new dimension in documentation through association of ideas. Science 122(3159):108–111
Gittelman M, Kogut B (2003) Does good science lead to valuable knowledge? Biotechnology firms and the evolutionary logic of citation patterns. Manage Sci 49(4):366–382
Hood L, Rowen L (2013) The human genome project: big science transforms biology and medicine. Genome Med 5:1–8
Huang M-H, Yang H-W, Chen D-Z (2015) Increasing science and technology linkage in fuel cells: A cross citation analysis of papers and patents. J informetrics 9(2):237–249
Jinek M, Chylinski K, Fonfara I, Hauer M, Doudna JA, Charpentier E (2012) A programmable dual-RNA–guided DNA endonuclease in adaptive bacterial immunity. Science 337(6096):816–821
KB M Specific enzymatic amplification of DNA in vitro: the polymerase chain reaction. Cold Spring Harbor Symp. Quant Biol, 51
Kwon S, Liu X, Porter AL, Youtie J (2019) Research addressing emerging technological ideas has greater scientific impact. Res Policy 48(9):103834
Lo S-c (2010) Scientific linkage of science research and technology development: A case of genetic engineering research. Scientometrics 82(1):109–120
Martin-Laffon J, Kuntz M, Ricroch AE (2019) Worldwide CRISPR patent landscape shows strong geographical biases. Nat Biotechnol 37(6):613–620
Meyer M (2000) Does science push technology? Patents citing scientific literature. Res Policy 29(3):409–434
Mullis KB (1994) The polymerase chain reaction, vol 41. Springer science & business media
Narin F (1994) Patent bibliometrics. Scientometrics 30(1):147–155
Narin F, Hamilton KS, Olivastro D (1995) Linkage between agency-supported research and patented industrial technology. Res Evaluation 5(3):183–187
Narin F, Hamilton KS, Olivastro D (1997) The increasing linkage between US technology and public science. Res Policy 26(3):317–330
Narin F, Noma E (1985) Is technology becoming science? Scientometrics 7(3):369–381
Schmoch U (1997) Indicators and the relations between science and technology. Scientometrics 38:103–116
Shapira P, Wang J (2010) Follow the money. Nature 468(7324):627–628
Tijssen RJ (2001) Global and domestic utilization of industrial relevant science: patent citation analysis of science–technology interactions and knowledge flows. Res Policy 30(1):35–54
Van Looy B, Zimmermann E, Veugelers R, Verbeek A, Mello J, Debackere K (2003) Do science-technology interactions pay off when developing technology? Scientometrics 57(3):355–367
Van Raan AF (2017) Patent citations analysis and its value in research evaluation: A review and a new approach to map technology-relevant research. J Data Inform Sci 2(1):13–50
Van Vianen B, Moed H, Van Raan A (1990) An exploration of the science base of recent technology. Res Policy 19(1):61–81
Watson JD (1990) The human genome project: past, present, and future. Science 248(4951):44–49

The authors declare no competing interests.

Appendix.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Is Science Driving Technology? An Investigation in the Field of Breast Cancer: Analysis of Polymerase Chain Reaction and CRISPR/Cas9

Status:

Version 1

Abstract

Figures

1 Introduction

2 Methods

2.1 Science: Data source and data selection of scientific publications

2.2 Patents data selection

3 Results

3.1 Results for Polymerase Chain Reaction

3.2 Results for CRISPR/Cas9

3.3 Nonpatent references

4 Discussion

4.1 Polymerase Chain Reaction

4.2 Clustered Regularly Interspaced Short Palindromic Repeats

4.3 Country Analysis

4.4 Non-Patent References

5 Conclusion

References

Additional Declarations

Supplementary Files

Status:

Version 1