There is tremendous scientific and technological potential in the field of Proteomics. Human diseases involve proteins, yet our current level of understanding of the proteome is incomplete, with major gaps in knowledge about the composition, localization and dynamics of proteins. Identifying what are the next key breakthroughs in protein-level chemistry and biology is of high interest to government and foundation stakeholders of biomedical research and companies (including their suppliers) as many aspire to see genomics-level impact from proteomics this decade.
However the big question remains: where is this field headed in the next 5-10 years? what gaps in our knowledge remain and what new approaches and technologies are likely to be deployed to address these gaps? In order to answer these questions, we applied the Consilience.ai engine that analyze a corpus of 29 journals in PubMed across about 25K publications on this topic. Based on this corpus, we applied the Consilience’s explainable, information-theoretic AI approaches like LDA (Latent Dirichlet Allocation) and Network Analysis to map macro-level trends in proteomics by identify groupings of similar words – called ideas – and analyze the growth and development of 94 ideas across the 2000s and 2010s through to 2023. In addition to individual ideas, our AI approach is also designed to reveal hidden relationships between ideas and concepts in large datasets.
From the network of these 91 ideas, we prioritized to the top 4 ideas as well as relationships between ideas using several models such as the Popularity, Network Connection, and Rate of Rise for individual. These identified the most stable ideas that will transform our ability to measure and understand proteins in the coming decade.
Here are the 4 top ideas (and relationships) identified:
Compositional Proteomics: A dramatic shift is happening as the key priorities are shifting from bottom up proteomics (by breaking apart proteins into peptides and then analyzing components) to top-down proteoform analysis (ie analying intact proteins. This shift is party driven by the understanding that bottom up proteomics does not reveal the various isoforms (or the proteoforms) of the proteomics and this is where the focus is going to be over the next several years – ie gaining a deeper understanding of structure and function of various proteoforms. Rate of Rise metric: 80%
Single Cell Proteomics: Recent understanding from the genomics has revealed that analyzing at a bulk level often masks important insights of rare events happening inside individual cells. This has triggered growth of single cell genomics field over the past decade. A similar realization is driving the need for single cell proteomics
Spatial Proteomics: Increasing evidence in recent years to demonstrate that specific locations of cells with specific protein markers inside tissues (e.g. cancer biopsies) can be prognostic and also predict therapy response.
Protacs: Proteolysis targeting chimera (ProTAC) as an approach for small molecule mediated removal of proteins that are undesirable, drug targets or truncated and dysfunctional. PROTACs is following in footsteps of other drug discovery techniques including the use of Kinase inhibitors, tools, compounds, and drug leads in the kinome. Key idea synergies are in the top decile, proven to be high performing and stable areas of opportunity
