Identify novel elements of knowledge with word embedding.

As novelty is a core value in science, a reliable approach to measuring the novelty of scientific documents is critical. Previous novelty measures however had a few limitations. First, the majority of previous measures are based on recombinant novelty concept, attempting to identify a novel combination of knowledge elements, but insufficient effort has been made to identify a novel element itself (element novelty). Second, most previous measures are not validated, and it is unclear what aspect of newness is measured. Third, some of the previous measures can be computed only in certain scientific fields for technical constraints. This study thus aims to provide a validated and field-universal approach to computing element novelty. We drew on machine learning to develop a word embedding model, which allows us to extract semantic information from text data. Our validation analyses suggest that our word embedding model does convey semantic information. Based on the trained word embedding, we quantified the element novelty of a document by measuring its distance from the rest of the document universe. We then carried out a questionnaire survey to obtain self-reported novelty scores from 800 scientists. We found that our element novelty measure is significantly correlated with self-reported novelty in terms of discovering and identifying new phenomena, substances, molecules, etc. and that this correlation is observed across different scientific fields.

Role of machine and organizational structure in science.

The progress of science increasingly relies on machine learning (ML) and machines work alongside humans in various domains of science. This study investigates the team structure of ML-related projects and analyzes the contribution of ML to scientific knowledge production under different team structure, drawing on bibliometric analyses of 25,000 scientific publications in various disciplines. Our regression analyses suggest that (1) interdisciplinary collaboration between domain scientists and computer scientists as well as the engagement of interdisciplinary individuals who have expertise in both domain and computer sciences are common in ML-related projects; (2) the engagement of interdisciplinary individuals seem more important in achieving high impact and novel discoveries, especially when a project employs computational and domain approaches interdependently; and (3) the contribution of ML and its implication to team structure depend on the depth of ML.

Measuring novelty in science with word embedding.

Novelty is a core value in science, and a reliable measurement of novelty is crucial. This study proposes a new approach of measuring the novelty of scientific articles based on both citation data and text data. The proposed approach considers an article to be novel if it cites a combination of semantically distant references. To this end, we first assign a word embedding-a vector representation of each vocabulary-to each cited reference on the basis of text information included in the reference. With these vectors, a distance between every pair of references is computed. Finally, the novelty of a focal document is evaluated by summarizing the distances between all references. The approach draws on limited text information (the titles of references) and publicly shared library for word embeddings, which minimizes the requirement of data access and computational cost. We share the code, with which one can compute the novelty score of a document of interest only by having the focal document's reference list. We validate the proposed measure through three exercises. First, we confirm that word embeddings can be used to quantify semantic distances between documents by comparing with an established bibliometric distance measure. Second, we confirm the criterion-related validity of the proposed novelty measure with self-reported novelty scores collected from a questionnaire survey. Finally, as novelty is known to be correlated with future citation impact, we confirm that the proposed measure can predict future citation.

Impact of Ph.D. training: a comprehensive analysis based on a Japanese national doctoral survey.

Ph.D. training in academic labs offers the foundation for the production of knowledge workers, indispensable for the modern knowledge-based society. Nonetheless, our understanding on Ph.D. training has been insufficient due to limited access to the inside of academic labs. Furthermore, early careers of Ph.D. graduates are often difficult to follow, which makes the evaluation of training effects challenging. To address these limitations, this study aims to illustrate the settings of Ph.D. training in academic labs and examine their impact on several training outcomes, drawing on a national survey of a cohort of 5000 Ph.D. graduates from Japanese universities. The result suggests that a supervising team structure as well as the frequency of supervision, contingent to a few contextual factors, determine the Ph.D. graduates' career decisions, performance, and degrees of satisfaction with the training programs.

Contribution of structural biology to clinically validated target proteins.

We identified six groups of diseases expected to cause serious future health issues on the basis of a WHO report. Approved drugs for these diseases were associated with 409 target proteins; however, the percentage of selected proteins with full-length structural information deposited in the Protein Data Bank (PDB) was only 9.8%. The reason for the low percentage may be as a result of a disproportionate number of intractable proteins with multiple transmembrane regions and variable, or undefined glycosylation patterns, which impede protein preparation and crystallization, in such druggable proteins. We stress the importance of structural analysis of proteins, especially approved-drug target proteins, and the development of new methods to enable structural analyses of presently intractable proteins. In addition, we present an overview of large structural biology projects.

Effect of mergers and acquisitions on drug discovery: perspective from a case study of a Japanese pharmaceutical company.

The pharmaceutical industry has experienced intermittent waves of mergers and acquisitions (M&As) since the 1980s and recently appeared to be in yet another wave. Previous studies indicated rather negative impacts of consolidation on research and development, suggesting that they do not necessarily lead to long-term reinforcement of research capabilities, although they may enrich the drug pipeline in the short term. However, recent studies have implied a positive side in terms of knowledge-base transfer. Further micro-organizational studies suggested that scientists learned new knowledge and approaches from partner scientists and improved their performance and innovation. These findings imply that measures for the scientist-level integration after M&As would reinforce fundamental research capabilities in the long term.

Identification of a C-terminal region that is required for the nuclear translocation of ERK2 by passive diffusion.

Extracellular signal-regulated kinase 2 (ERK2) is located in the cytoplasm of resting cells and translocates into the nucleus upon extracellular stimuli by active transport of a dimer. Passive transport of an ERK2 monomer through the nuclear pore is also reported to coexist. We attempted to characterize the cytoplasmic retention and nuclear translocation of fusion proteins between deletion and site-directed mutants of ERK2 and green fluorescent protein (GFP). The overexpressed ERK2-GFP fusion protein is usually localized to both the cytoplasm and the nucleus unless a cytoplasmic anchoring protein is coexpressed. Deletion of 45 residues, but not 43 residues, from the C terminus of ERK2 prevented the nuclear distribution of the ERK2-GFP fusion protein. Substitution of a part of residues 299-313 to alanine residues also prevented the nuclear distribution of the ERK2-GFP fusion protein without abrogation of its nuclear active transport. These observations may indicate that the passive diffusion of ERK2 into the nucleus is not simple diffusion but includes a specific interaction process between residues 299-313 and the nuclear pore complex and that this interaction is not required for the active transport. We also showed that substitution of Tyr(314) to alanine residue abrogated the cytoplasmic retention of the ERK2-GFP fusion protein by PTP-SL but not by MEK1.