Statistics of Article Publications and Assessment of Future Growth Trends
The number of articles published in a scientific field at different stages can reflect the development prospects of the field. After the above search strategy and literature collection, a total of 945 literatures were retrieved in this study, of which 813 original articles (86.03%) and 132 reviews (13.97%) were identified, respectively (Figure 2A). Online Analysis Platform of Bibliometrics (http://bibliometric.com/) was used to count the actual number of published articles in this field each year (Figure 2B). It can be seen intuitively from the histogram that although the number of articles declined at a few time points, the number of published articles related to this field showed an overall upward trend and reached a peak in 2017, with a total of 85 papers published, accounting for 8.99% of the total number of publications. What caught our attention was that the number of published articles in this field was 0 in the early period (2001-2006) and started to grow rapidly after 2007. The number of papers this year,2022, has reached 21 as of June 15, 2022. On the whole, with the in-depth study of "metabolism affects diseases" in modern medicine, the role of Gln metabolism in diabetes has gradually attracted extensive attention of scholars, and the interaction and progress of the two have become a hot topic of research. In recent years, more and more studies have gradually revealed the pathogenesis and targets of Gln metabolism in diabetes[12,25,26].
In order to identify research countries/regions that have made outstanding contributions to this field, we analyzed the number of articles published in different countries/regions. The top 10 countries/regions in the total number of articles published in the past 20 years were shown in the area chart (Figure 2C). The United States was the breakthrough in this field, and the number of articles published was still in a steady growth stage. Furthermore, since 2013, in addition to the United States, China, the United Kingdom and Germany have also begun to dominate the number of articles published in this research field, and China has been in a rapid growth phase since 2017. What we can expect is that in the near future, the number of published articles in China will further increase. The interconnection and cooperation between different countries are shown in Figure 2D. The thickness of the line between different countries indicates the strength of cooperation between two countries. It can be seen that the United States currently cooperates most closely with China, Spain, France, Canada and Germany. But overall, the linkages and cooperation between the various countries need to be further strengthened.
Analyze the Cooperation between Countries/Regions, or between Institutions
We used CiteSpace to analyze the mutual cooperation between countries/regions, or between institutions, in order to visually demonstrate the close relationship of cooperation in the fields of Gln metabolism and diabetes. A total of 59 countries/regions (Figure 2D) and 1302 institutions contributed to this research field. Table 1 summarizes the top 20 high-yield institutions according to publications counts. Vanderbilt University was the institutions with the largest number of articles in the field of Gln metabolism and diabetes (90, 95.23 %), followed by Yale University (90, 95.23 %), University of Texas Southwestern Medical Center (79, 83.59 %), University of Pennsylvania (76, 80.42 %), Emory University (57, 60.32 %) and University of Michigan (57, 60.32 %). As we have shown, it was no surprise that the United States was at the top of the list in terms of the number of publications, and the field advantage was greatly exploited, further strengthening the academic influence of the United States in this field.
Table 1. Top 20 institutions with the most publications related to diabetes and Gln metabolism AND diabetes and glutaminolysis research.
Ranking
|
Location
|
Institution
|
Count
|
Percentage (% of 945)
|
Average Citations per Item
|
Sum of Times Cited
|
1
|
United States
|
Vanderbilt Univ
|
90
|
95.23
|
1.77
|
159
|
2
|
United States
|
Yale Univ
|
79
|
83.59
|
2.94
|
232
|
3
|
United States
|
Univ Texas SW Med Ctr Dallas
|
76
|
80.42
|
10.25
|
779
|
4
|
United States
|
Univ Penn
|
72
|
76.19
|
4.18
|
301
|
5
|
United States
|
Emory Univ
|
57
|
60.32
|
0.91
|
52
|
6
|
United States
|
Univ Michigan
|
57
|
60.32
|
0.67
|
38
|
7
|
United States
|
Harvard Med Sch
|
53
|
56.08
|
0.85
|
45
|
8
|
United States
|
Univ Louisville
|
43
|
45.51
|
3.86
|
166
|
9
|
Netherlands
|
Leiden Univ
|
43
|
45.51
|
0.21
|
9
|
10
|
United States
|
Univ Florida
|
42
|
44.44
|
2.86
|
120
|
11
|
United States
|
Duke Univ
|
36
|
38.09
|
1.25
|
45
|
12
|
Denmark
|
Univ Copenhagen
|
34
|
35.98
|
1.62
|
55
|
13
|
United States
|
Childrens Hosp Philadelphia
|
31
|
32.81
|
7.11
|
220
|
14
|
United States
|
Univ Calif San Diego
|
31
|
32.81
|
2.52
|
78
|
15
|
United States
|
Johns Hopkins Univ
|
31
|
32.81
|
1.68
|
52
|
16
|
United States
|
Univ Kentucky
|
31
|
32.81
|
1.13
|
35
|
17
|
United States
|
Univ Alabama Birmingham
|
30
|
31.75
|
1.21
|
36
|
18
|
United States
|
Univ Washington
|
30
|
31.75
|
0.53
|
16
|
19
|
United States
|
Univ Calif Davis
|
29
|
30.69
|
2.14
|
62
|
20
|
United States
|
MIT
|
28
|
29.63
|
10.96
|
307
|
Analysis by the Authors Who Contributed the Most
The higher the number of articles an author published, the greater the level of scholarly activity and contribution to the research field. Judging by the perspective of publication counts (Table 2), the most productive author with 20 publications in this field was identified as Nissim. I, Weiner. ID (18 publications) and DeBerardinis. RJ (17 publications) ranked second and third, respectively.
Authors working on different research topics have unique professional characteristics in this field, and cross-collaboration can also promote interaction and quantitative output in this research field. In addition, analyzing an author and his co-authors helps our researchers understand existing partnerships, communicate further, and develop potential collaborative themes[14]. As shown in Figure 4, a visual map of the collaboration between authors is analyzed and synthesized by the VOS viewer. As the figure shows, several concentrated research groups have been created in this field, with different colors representing clusters of closely collaborating authors. Groups in each collection are connected by one or two core authors with outstanding contributions, such as Nissim. I, Weiner. ID, DeBerardinis. RJ, and Rothman. DL. Overall, the most prominent point is the weak mutual cooperation between the different ensemble groups, indicating that the exchanges and connections in this field are not well developed and need our further efforts. Likewise, CiteSpace was also used to identify mutual collaborations between different authors, showing results consistent with those described above, as shown in Supplementary Figure 2.
Table 2. Top 10 authors with the most publications related to diabetes and Gln metabolism AND diabetes and glutaminolysis research.
Ranking
|
Author
|
Total publications
|
Citations
|
1
|
Nissim. I
|
20
|
151
|
2
|
Weiner. ID
|
18
|
104
|
3
|
DeBerardinis. RJ
|
17
|
333
|
4
|
Verlander. JW
|
16
|
89
|
5
|
Rothman. DL
|
16
|
46
|
6
|
Lee. HW
|
14
|
71
|
7
|
Behar. KL
|
14
|
45
|
8
|
Stephanopoulos. G
|
12
|
134
|
9
|
Metallo. CM
|
12
|
108
|
10
|
Mason. GF
|
11
|
38
|
Analysis of the Higher-Impact and Higher-Cited Journals
For many years, the influence of journal publications has always been an important tool and carrier for scientists and researchers in various research fields to conduct scientific communication, reference and citation. Analyzing valid information from journal sources helps researchers quickly find the most suitable journals for their publications. In addition to this, presenting a field of research in international peer-reviewed journals is an important part of establishing effective scholarly communication[14,18]. Articles on Gln metabolism and diabetes research were published in 379 different journal publications, many of which were specialized academic journals. As shown in Supplementary Table 1, the three most influential journals in terms of number of publications were Journal of Biological Chemistry (31, 3.28 %), Plos One (29, 3.07 %), and Cell Metabolism (20, 2.12 %).
The influence of journals depends not only on the number of articles published, but also on the number of times they are co-cited in the field of research[24]. Table 3 lists information on the top 20 most cited journals. Among them, the top five journals with the most citations are Proceedings of the National Academy of Sciences of the United States of America (174 citations), Nature (126 citations), Cell Metabolism (123 citations), American Journal of Physiology-Renal Physiology (72 citations) and Journal of Biological Chemistry (57 citations). The above results showed that these journal publications have published a large number of high-profile research results, which attracted great attention and reference orientation by researchers interested in the field.
Furthermore, a journal's IF measures the average frequency with which articles published by the journal are cited in a given year, and is used to assess the relative importance of different journals in the same research field. In 2006, Garfield proposed the concept of journal IF, introduced the history and significance of IF, which was designed to measure a journal's 2-year moving average citation volume and reach[27]. Among the top 20 most total citations academic journals, Nature (IF = 49.962) has the highest IF, followed by Cell Metabolism (IF = 27.287), Journal of Clinical Investigation (IF = 14.808), Cancer research (IF = 12.701) and Molecular Systems Biology (IF = 11.429). Journal Citation Reports (JCR) is an effective tool for systematic and objective evaluation of authoritative journals in the world and was published annually. If we wanted to understand the importance and influence of a certain academic journal in a certain research field, JCR was undoubtedly an ideal and indispensable tool[28]. JCR used the citation information provided by the Science Citation Index (SCI) and the Social Science Citation Index (SSCI) to conduct statistical analysis, builded a unique database that can be used for academic journal analysis and evaluation. It also divided journals that belong to the same category into four equal parts, among which the top 25% attributed to Q1 and the top 25 50% being Q2, and so forth[29]. As can be seen from Table 3 that 70 % of journals belong to Q1. It is foreseeable that these journals may publish more high-quality academic research in the future, further improving the influence and development prospects of this field. Meanwhile, in this work, we performed a co-citation analysis of journals by using the VOS viewer to explore the connections between different journals published in this research area from 2001 to 2022 (Supplementary Figure 3). It also screened out the indispensable professional journals for researchers in this field.
Table 3. Top 20 journals with the most total citations in the field of diabetes and Gln metabolism AND diabetes and glutaminolysis research.
Ranking
|
Journal Name
|
Citations
|
Count
|
Percentage (% of 945)
|
IF# (2021)
|
H-index
|
JCR& (2021)
|
1
|
Proceedings of the National Academy of Sciences of the United States of America
|
174
|
14
|
1.48
|
11.205
|
771
|
Q1
|
2
|
Nature
|
126
|
6
|
0.63
|
49.962
|
652
|
Q1
|
3
|
Cell Metabolism
|
123
|
20
|
2.12
|
27.287
|
266
|
Q1
|
4
|
American Journal of Physiology-Renal Physiology
|
72
|
15
|
1.59
|
3.377
|
169
|
Q2
|
5
|
Journal of Biological Chemistry
|
57
|
31
|
3.28
|
5.157
|
513
|
Q1
|
6
|
Oncogene
|
43
|
3
|
0.32
|
9.867
|
342
|
Q1
|
7
|
Journal of Cerebral Blood Flow and Metabolism
|
34
|
10
|
1.06
|
6.200
|
193
|
Q1
|
8
|
Amino Acids
|
29
|
6
|
0.63
|
3.520
|
118
|
Q2
|
9
|
Cancer Research
|
29
|
9
|
0.95
|
12.701
|
449
|
Q1
|
10
|
Plos One
|
26
|
29
|
3.07
|
3.240
|
332
|
Q1
|
11
|
Nutrition
|
26
|
9
|
0.95
|
4.008
|
142
|
Q2
|
12
|
Journal of Clinical Investigation
|
26
|
4
|
0.42
|
14.808
|
488
|
Q1
|
13
|
Diabetologia
|
24
|
8
|
0.85
|
10.122
|
227
|
Q1
|
14
|
Diabetes
|
22
|
14
|
1.48
|
9.461
|
330
|
Q1
|
15
|
Current Opinion in Genetics & Development
|
21
|
2
|
0.21
|
5.578
|
189
|
Q2
|
16
|
Clinical Journal of the American Society of Nephrology
|
19
|
2
|
0.21
|
8.237
|
151
|
Q1
|
17
|
Journal of Neurochemistry
|
18
|
4
|
0.42
|
5.372
|
229
|
Q2
|
18
|
Journal of Clinical Endocrinology & Metabolism
|
17
|
10
|
1.06
|
5.958
|
353
|
Q1
|
19
|
Nutrition & Metabolism
|
16
|
3
|
0.32
|
4.169
|
84
|
Q3
|
20
|
Molecular Systems Biology
|
13
|
1
|
0.11
|
11.429
|
148
|
Q1
|
#IF, impact factor. &JCR, journal citation reports. Q, quartile in category.
Analysis of Co-Citation References
Reference co-citation analysis refers to the documents that are cited by other documents at the same time as this article, indicating the close relationship between the two documents. It is also a valuable technique for assessing the evolution of a research field and tracking the frontiers of development, revealing the authoritativeness of research in the field and the great contributions of authors[30]. The analysis was run in CiteSpace software with the time slice set to one year and the time span set to 2001 to 2022 (Supplementary Figure 4). Figure 5A presents the visualized network map of cited references. By using the clustering function, the entire network graph can be aggregated into several different classes, and the largest 19 clusters are extracted using the log-likelihood ratio algorithm (LLR). Figure 5A shows them with different analogies, including sirtuins (cluster #0), stable isotope-resolved metabolomics (cluster #1), hepatic encephalpathy (cluster #2), l-glutamine (cluster #3), osteoblast (cluster #4), glutamate dehydrogenase (cluster #6), type 2 diabetes (cluster #7), diseases (cluster #8), lung cancer (cluster #9), futile cycles (cluster #10), and so on. In addition, two significant and important parameters for evaluating the aggregated community structure are the modularity value (Q-value) and the mean contour value (S-value), respectively, when Q > 0.3 and S > 0.5 indicate significant clustering[14,31]. In our study, the Q-value was 0.906, indicating the rationality of this network graph. The mean S-value was 0.95, with all S-values greater than 0.9275 for the #0 - #26 metacommunities, indicating good homogeneity of these metacommunities.
Furthermore, the network graph of timeline is a visualization method combining clustering analysis and time slicing technology, which is used to study the knowledge evolution path of a specific discipline domain[15]. Figure 5B depicts a timeline network map of co-citations across different clusters. All cited references can be clearly displayed. The bold timeline indicates that a cluster topic was a hot topic during this period. Tree-rings of different sizes on the timeline represent high-quality articles with high citation frequency[32]. We found that in this research on Gln metabolism in diabetes progression, stable isotope-resolved metabolomics has been a hotspot since 2006, reaching its peak moment in 2012, and has been in high interest. l-glutamine was an emerging research field in 2011 and has attracted increasing attention recently. Studies on glutamate dehydrogenase first appeared in 2002 and stabilized in the future. To our interest, type 2 diabetes (cluster #7) is the most recent region identified and has been under high concern, suggesting that Gln metabolism may play an important and critical role in the progression of diabetes.
Analyze the Most Frequently Occurring Keywords
In addition to analyzing references, the analysis of keywords also represents the core and thematic content of a particular subject area, in order to reveal the content relevance of field research information and the keywords knowledge implied by feature items[33]. For scientometric, another commonly used method for identifying hot research topics and areas is keywords co-occurrence analysis, which can be used to analyze research trends and development directions, as well as to monitor the shift of research fronts and the emergence of emerging issues within a knowledge domain. In this study, keyword clusters with a minimum of 5 occurrences were generated by the VOS viewer, as shown in Figure 6A, and the size of each node represents the frequency of keyword occurrences. Four colors represent different four clusters, clusters with common characteristics and attributes are classified into the same color-coded cluster, represented by green, blue, red, and yellow, which revolved around “glutamine plus glucose”, “glutamine-metabolism”, “insulin-resistance”, and “metabolism”, respectively. After removing several meaningless and irrelevant keywords, we merged keywords with the same meaning, Figure 6B provides an overlay visualization of the network map for the most frequent keywords in this field. The frequency of occurrence of keywords is proportional to the size of nodes in the map, and the distance between two nodes indicates the closeness of the relationship between keywords. In addition to that, the thicker the line between two nodes, the more often they appear together. For example, in this study, "metabolism" appeared the most, followed by "glutamine". The thick line between "metabolism" and "glutamine-metabolism", "cancer", "diabetes" and "obesity" means that they appear at the same time. To analyze the changes in the number of different keywords over the past 20 years and to study the relationship between different keywords, this study used the Online Analysis Platform of Bibliometrics (http://bibliometric.com/) to create a bar chart, showed the frequency distribution of keywords (Figure 7A). It can be seen that since 2007, the keywords “glutamine”, “metabolomics” and “diabetes” have been co-occurring with each other and occupy a high frequency of studies. It is commonly held that regulation of Gln metabolism plays a critical role in the progression of diabetes. Thus, further research and elucidation of these molecular mechanisms of Gln metabolism may facilitate the discovery of new strategies and hopes for the treatment of diabetes. In addition, it is worth our attention and excitement that in the first half of 2022, the keyword “glutamine”, “metabolomics”, “metabolism”, “diabetes”, and “type 2 diabetes” showed multiple study frequencies, which also indicated that this topic has received more and more attention recently and was still the most popular research hotspot in recent years. The bar chart of changes in the number of extended keywords over the past 20 years was shown in Figure 7B.
Furthermore, the landscape representation of the keywords ensemble classes was analyzed and generated by CiteSpace software, presenting 17 ensemble swarm classes (Figure 8A), and the overlapping portion of the image indicates that this study appears and is cited in multiple swarm classes. Cluster #0 labeling the skeletal muscle, was the largest cluster, followed by expression (cluster #1), and nitric oxide (cluster #2). Apart from this, several research directions, including hyperaminoacidemia (cluster #4), insulin secretion (cluster #5), and cell (cluster #7) were the main topics since 2007. Meanwhile, tca cycle (cluster #11), arginine (cluster #12), and type 2 diabetes (cluster #13) has been a research hotspot in this field. In addition, the co-occurrence keywords timeline graph is shown in Figure 8B, which consists of 474 nodes and 1,872 links, representing the occurrence frequency and co-occurrence relationship of keywords.
Analyze References and Keywords Using Burst Detection
Burst detection, a powerful tool that captures the sharp increase in the popularity of a references or keywords over a specific period, is an algorithm developed by Kleinberg[34], and detected research in a field from macro to micro, from single to the evolutionary trend of diversification. This feature is an effective way to identify topics or concepts that have been discussed frequently over a period of time, as well as draw the attention of peer investigators to emerging concepts and future trends. Two properties of citation bursts are intensity and duration of state[15]. In our study, burst detection was used to extract key references and keywords. By using CiteSpace, Figure 9 illustrated the top 25 references with the strongest citation bursts during the period of 2001 to 2022. The blue part represents the time interval, and the red part represents the time period of the reference burst in this figure. Among them, the strongest reference for burst value (strength: 10.78) was published in the Nature and written by Metallo. CM et al.. In this study, they found that use and regulation of Gln metabolism in hypoxic cells. These results identify a critical role for oxygen in regulating carbon use to produce Acetyl coenzyme A (AcCoA) and support lipid synthesis in mammalian cells[35]. Furthermore, it also can be seen that in 2007, the first burst of co-cited reference began [36] and the most recent references with citation bursts appeared in 2020. A total of 5 references had a burst that lasted until 2022[37-39]. The references with citation bursts between 2010 and 2020 accounted for 88.00 %. Notably, although the burst of most references has ended, the burst of some references continues, which indicates that these research topics or key technologies have been receiving attention in recent years.
Detection and analysis of burst keywords can show the number of times keywords are frequently cited over a period of time[40,41], which can be considered as an indicator direction of cutting-edge research topics. Figure 10 shows the top 25 keywords with citation bursts, the red part represents the time interval of the burst. A stronger burst rate indicates more attention to the research topic. Over the past time, “type 2 diabetes” ranked first with the highest burst strength (strength: 7.69), followed by “proliferation” (strength: 5.01), “transport” (strength: 4.89), “rat brain” (strength: 4.31), and “pancreatic beta cell” (strength: 4.16). More importantly, “type 2 diabetes”, “risk”, “insulin sensitivity”, “tac cycle”, “individual”, and prevalence became the focus from 2019 until now, indicates that they are current research hot topics.