Independent Discovery Research Articles

A Historical Review of the Artificial Pollination of Vanilla planifolia: The Importance of Collaborative Research in a Changing World

The natural fertilization of Vanilla planifolia has long been a matter of speculation. Stingless (tribe Meliponini) and orchid bees (tribe Euglossini) are often cited as effective pollinators, but direct evidence is notoriously lacking. As with other food-deceptive orchids, the natural fruit-set in V. planifolia is low and does not occur spontaneously outside its native range in Middle America. Fruiting has, therefore, necessitated human intervention through artificial pollination. How Vanilla first came to be artificially pollinated is a controversial issue spurring egotism and nationalism. There are numerous texts offering differing versions of the historical events that led to the discovery of the artificial fertilization of V. planifolia and its propagation as a crop. Historical records show Vanilla was simultaneously being pollinated in several parts of the world. I argue that the suspected independent simultaneous discoveries made in Liège, Paris, Padua, the Dutch colony of Java (Indonesia), and the French insular colonies Réunion (Bourbon), Guadeloupe, and Martinique are not unconnected. I conclude that they can be traced back to a single discoverer whose feat was spread around the globe by a tight network of corresponding naturalists. This view contrasts with previous authors. Finally, current concerns regarding Vanilla pollination and production are addressed, highlighting the need for immediate actions to conserve the genetic diversity of the crop’s wild relatives to attenuate the effect of extreme climates in a changing world. A plea is made to shift the focus to Middle America, stimulating and supporting local research and conservation efforts and the allocation of funds from this billion-dollar industry.

CD44: A Stemness Driver, Regulator and Marker - all in one?

Although the concept of cancer stem cells is still controversial, previous studies have shown that blood cancers, as well as specific types of solid cancers such as colorectal cancer, rely on stem cells during the onset of tumor growth and further tumor development. Moreover, resistance to therapeutic treatment in leukemias such as acute myeloid leukemia and in colorectal cancer can be attributed to a small population of cells with stemness properties known as minimal residual disease. In this review, we look back on the discovery of cancer stem cells and the contribution of the findings in blood cancer to a parallel discovery in solid cancers. We focus on CD44 as a stem cell marker, both in blood cancers and in several types of solid cancers, particularly of the gastro-intestinal tract. This review highlights newly discovered molecular mechanisms of action of CD44 which indicate that CD44 has indeed a function in stemness, stem cell maintenance and drug resistance. We attempt here to make the link between the functions of CD44 isoforms in stemness and their involvement in specific steps of tumor growth and metastasis.

Machine Learning-Based Prediction of Neurodegenerative Disease in Patients With Type 2 Diabetes by Derivation and Validation in 2 Independent Korean Cohorts: Model Development and Validation Study.

Several machine learning (ML) prediction models for neurodegenerative diseases (NDs) in type 2 diabetes mellitus (T2DM) have recently been developed. However, the predictive power of these models is limited by the lack of multiple risk factors. This study aimed to assess the validity and use of an ML model for predicting the 3-year incidence of ND in patients with T2DM. We used data from 2 independent cohorts-the discovery cohort (1 hospital; n=22,311) and the validation cohort (2 hospitals; n=2915)-to predict ND. The outcome of interest was the presence or absence of ND at 3 years. We selected different ML-based models with hyperparameter tuning in the discovery cohort and conducted an area under the receiver operating characteristic curve (AUROC) analysis in the validation cohort. The study dataset included 22,311 (discovery) and 2915 (validation) patients with T2DM recruited between 2008 and 2022. ND was observed in 133 (0.6%) and 15 patients (0.5%) in the discovery and validation cohorts, respectively. The AdaBoost model had a mean AUROC of 0.82 (95% CI 0.79-0.85) in the discovery dataset. When this result was applied to the validation dataset, the AdaBoost model exhibited the best performance among the models, with an AUROC of 0.83 (accuracy of 78.6%, sensitivity of 78.6%, specificity of 78.6%, and balanced accuracy of 78.6%). The most influential factors in the AdaBoost model were age and cardiovascular disease. This study shows the use and feasibility of ML for assessing the incidence of ND in patients with T2DM and suggests its potential for use in screening patients. Further international studies are required to validate these findings.

Radio Afterglows from Tidal Disruption Events: An Unbiased Sample from ASKAP RACS

Late-time (∼a year) radio follow-up of optically discovered tidal disruption events (TDEs) is increasingly resulting in detections at radio wavelengths, and there is growing evidence for this late-time radio activity to be common to the broad class of subrelativistic TDEs. Detailed studies of some of these TDEs at radio wavelengths are also challenging the existing models for radio emission. Using all-sky multiepoch data from the Australian Square Kilometre Array Pathfinder (ASKAP), taken as a part of the Rapid ASKAP Continuum Survey (RACS), we searched for radio counterparts to a sample of optically discovered TDEs. We detected late-time emission at RACS frequencies (742–1032 MHz) in five TDEs, reporting the independent discovery of radio emission from TDE AT 2019ahk and extending the time baseline out to almost 3000 days for some events. Overall, we find that at least 22−11+15% of the population of optically discovered TDEs has detectable radio emission in the RACS survey, while also noting that the true fraction can be higher given the limited cadence (two epochs separated by ∼3 yr) of the survey. Finally, we project that the ongoing higher-cadence (∼2 months) ASKAP Variable and Slow Transients survey can detect ∼20 TDEs in its operational span (4 yr), given the current rate from optical surveys.

Pentagonal Symmetry in Aperiodic Cyclic Multiply Twinned Nano- and Mesodiamonds.

This Review provides the first comprehensive overview of the structure and properties of exotic sp3-crystalline aperiodic cyclic multiply twinned nano- (10-100 nm) and meso- (up to 1 mm) diamond particles (MTPs) exhibiting pentagonal symmetry. It spans their independent experimental discoveries (1963, 1964, 1972, and 1983) and theoretical structural insights (1993) to recent advancements. The Review focuses on high-symmetry MTPs formed by the fusion of multiple cubic diamond fragments through [111] facets. The sp3 diamond lattice of individual fragments offers a vast range of MTP varieties. These particles are shown to be a special case of aperiodic crystalline solids with limited dimensions and rotational symmetry, leading to a breakdown of translational invariance. Detailed mathematical analysis of the MTPs' lattices highlights the crucial role of central cores in determining the symmetry and effective dimensions of these structures. Both structural and kinetic aspects of the formation mechanisms of pentagonal diamond particles are considered, revealing the main role of embryo seeds (low fullerenes, polyhexacyclo[5.5.1.12,6.18,12.03,11.05,9]pentadecane (C15), and polyheptacyclo[5.5.1.12,7.19,14.03,13.06,11]octadecane (C18)) in determining the MTPs' symmetry and structure. The effective dimensions of multiply twinned diamonds are shown to be limited by structural stress caused by the mismatch of perfect pentagonal and tetrahedral dihedral angles. The extraordinary mechanical and electronic properties of multiply twinned diamonds are discussed, highlighting that hexagonal interfaces between cubic diamond fragments may determine exceptional ultrahard and quantum characteristics. The MTP X-ray diffraction spectra reveal clear pentagonal patterns with single (diamond decahedrons or dodecahedrons) or ten (diamond icosahedra) central 5-fold axes. A comparative analysis of experimental structural data and simulations at different theoretical levels demonstrates a perfect correspondence of theoretical models with crystalline lattices.

Reproducible Sex Differences in Personalized Functional Network Topography in Youth.

A key step towards understanding psychiatric disorders that disproportionately impact female mental health is delineating the emergence of sex-specific patterns of brain organization at the critical transition from childhood to adolescence. Prior work suggests that individual differences in the spatial organization of functional brain networks across the cortex are associated with psychopathology and differ systematically by sex. We aimed to evaluate the impact of sex on the spatial organization of person-specific functional brain networks. We leveraged person-specific atlases of functional brain networks defined using nonnegative matrix factorization in a sample of n = 6437 youths from the Adolescent Brain Cognitive Development Study. Across independent discovery and replication samples, we used generalized additive models to uncover associations between sex and the spatial layout ("topography") of personalized functional networks (PFNs). Next, we trained support vector machines to classify participants' sex from multivariate patterns of PFN topography. Finally, we leveraged transcriptomic data from the Allen Human Brain Atlas to evaluate spatial correlations between sex differences in PFN topography and gene expression. Sex differences in PFN topography were greatest in association networks including the fronto-parietal, ventral attention, and default mode networks. Machine learning models trained on participants' PFNs were able to classify participant sex with high accuracy. Brain regions with the greatest sex differences in PFN topography were enriched in expression of X-linked genes as well as genes expressed in astrocytes and excitatory neurons. Sex differences in PFN topography are robust, replicate across large-scale samples of youth, and are associated with expression patterns of X-linked genes. These results suggest a potential contributor to the female-biased risk in depressive and anxiety disorders that emerge at the transition from childhood to adolescence.

Robert K. Merton and Innovation-Related Topics: Three Theoretical Ideas

As in other disciplines, such as organization science, sociology, and economics, there are good reasons for scholars in innovation and entrepreneurship research to revisit the classics within and outside their own research fields. Robert K. Merton (1910–2003) is not usually explicitly connected to innovation research, but to fields such as the sociology of science and knowledge. However, his scholarly productions include texts relevant to innovation and entrepreneurship. In 1935, he published on invention and innovation in American industry, and in 1938, he put forward the idea of innovation as deviant behavior or nonconformist adaptive behavior (Merton 1938a). In 1945, he began writing about the concept of serendipity, which he later discussed several times. This paper discusses how theoretical insights from Merton may inspire entrepreneurship, creativity, and innovation research. The discussion is framed around Merton’s innovation-related ideas and their relevance today, based on 20–25 Merton publications from 1934 to 2004 drawn from his larger scholarly production. The paper extracts three main examples of Merton’s theoretical ideas and insights. The first is his interest in simultaneous independent discoveries (Merton 1961a; 1961b), which has later been seen as a zeitgeist perspective on (scientific) creativity. The second is his view of innovation as deviant behavior and an adaptive response (Merton 1938), which is relevant to, among others, discussion of the dark side of innovation. The third is his interest in the idea of serendipity in scientific discovery (Merton 1945; 1948a), which is relevant today in relation to entrepreneurial opportunity and accidental innovation.

Another Independent Discovery of the 1946 Eruption of the Recurrent Nova T Coronae Borealis

One of the earliest independent discoveries of the 1946 eruption of the recurrent nova T Coronae Borealis was by British amateur astronomer, Michael Woodman, on 1946 February 9 at ∼01:45 UT when the nova was magnitude ∼2.0. Though credited as an independent discovery at the time, only scant details appeared in contemporary reports. The circumstances of Woodman’s discovery are told from Woodman’s contemporaneous notes and his own recollections.

MolCFL: A personalized and privacy-preserving drug discovery framework based on generative clustered federated learning

In today’s era of rapid development of large models, the traditional drug development process is undergoing a profound transformation. The vast demand for data and consumption of computational resources are making independent drug discovery increasingly difficult. By integrating federated learning technology into the drug discovery field, we have found a solution that both protects privacy and shares computational power. However, the differences in data held by various pharmaceutical institutions and the diversity in drug design objectives have exacerbated the issue of data heterogeneity, making traditional federated learning consensus models unable to meet the personalized needs of all parties. In this study, we introduce and evaluate an innovative drug discovery framework, MolCFL, which utilizes a multi-layer perceptron (MLP) as the generator and a graph convolutional network (GCN) as the discriminator in a generative adversarial network (GAN). By learning the graph structure of molecules, it generates new molecules in a highly personalized manner and then optimizes the learning process by clustering federated learning, grouping compound data with high similarity. MolCFL not only enhances the model’s ability to protect privacy but also significantly improves the efficiency and personalization of molecular design. MolCFL exhibits superior performance when handling non-independently and identically distributed data compared to traditional models. Experimental results show that the framework demonstrates outstanding performance on two benchmark datasets, with the generated new molecules achieving over 90% in Uniqueness and close to 100% in Novelty. MolCFL not only improves the quality and efficiency of drug molecule design but also, through its highly customized clustered federated learning environment, promotes collaboration and specialization in the drug discovery process while ensuring data privacy. These features make MolCFL a powerful tool suitable for addressing the various challenges faced in the modern drug research and development field.

Phylogeny and bioprospecting: The diversity of medicinal plants used in cancer management

Societal Impact StatementAs the second‐leading cause of mortality worldwide, cancer is a major focus of drug discovery research. Traditional plant knowledge can guide the search for undiscovered compounds, but the efficacy of this approach for cancer, a highly complex disease affecting diverse tissues, is unknown. We investigated the patterns underlying plant selection for cancer treatment globally, finding certain lineages are repeatedly targeted. While this indicates therapeutic value, their relatedness with plants used for unrelated ethnobotanical uses suggests that plants are probably selected to treat cancer‐associated symptoms, rather than addressing tumour growth. Careful re‐examination and scoring of ethnobotanical reports may make the prediction of lineages for drug discovery more informative.Summary Cancer is a highly diverse disease and as the second‐leading cause of death worldwide is a focus of drug discovery research. Natural products have been shown to be a useful source of novel molecules for treating cancer. It is likely there are many plants with undiscovered molecules of therapeutic value, however identifying new leads from the vast diversity of plants is very challenging. Traditional knowledge might inform bioprospecting by predicting lineages of plants rich in therapeutically useful molecules. We characterise the phylogenetic diversity of plants used in traditional cancer management using a comprehensive genus‐level phylogeny of angiosperms, and a list of 597 genera used globally to treat different cancers. We phylogenetically predict which lineages may have elevated potential for drug discovery and assess the quality of the prediction. We demonstrate the independent and repeated targeting of specific lineages of plants by different peoples in different parts of the world. However, the lineages we report here as rich in plants used in traditional cancer management coincide with those for other ethnobotanical applications and contain few plants with proven anti‐cancer activity. That the same lineages are used to treat different cancers is suggestive of independent discovery of therapeutic value. However, it is likely that the traditional knowledge explored here is shaped by the selection of plants conferring milder effects for treating wider symptoms, such as tiredness or nausea, rather than for halting tumour growth. Accurate prediction of useful plant lineages for cancer management requires more nuanced information than is commonly provided in ethnobotanical records.

Gustave-Adolphe Hirn, the mechanical equivalent of heat, and the conservation of energy

ABSTRACT Alsatian engineer Gustave-Adolphe Hirn is best known to historians of science for his experimental determination of the mechanical equivalent of heat, first published in 1855. Since the 1840s, that equivalent has been closely associated with the conservation of energy, indeed often conflated with it. Hirn was one of Thomas Kuhn’s twelve ‘pioneers’ whose work he deemed relevant to the ostensible ‘simultaneous discovery’ of energy conservation. Yet Hirn never wholeheartedly embraced energy conservation. After reviewing his experimental work, his philosophical reflections, and his response to developments in heat theory, this article identifies three factors as having played central roles in this regard: Hirn’s deepest concerns were ontological, not energetic; none of the concepts basic to his natural philosophy were appropriate stand-ins for a quantitatively conserved energy; and, accurately reflecting the most common interpretation of the principle of the conservation of energy from the 1860s on – as exemplified already by Helmholtz in 1847 – Hirn associated its corpuscular-mechanical underpinning with the despiritualizing materialism he saw as dominating contemporary science. Hence although Hirn’s natural philosophy embraced sentiments quite in the spirit of the conservation of energy, he never explicitly subscribed to that principle.

Triumphs and Challenges of Natural Product Discovery in the Postgenomic Era.

Natural products have played significant roles as medicine and food throughout human history. Here, we first provide a brief historical overview of natural products, their classification and biosynthetic origins, and the microbiological and genetic methods used for their discovery. We also describe and discuss the technologies that revolutionized the field, which transitioned from classic genetics to genome-centric discovery approximately two decades ago. We then highlight the most recent advancements and approaches in the current postgenomic era, in which genome mining is a standard operation and high-throughput analytical methods allow parallel discovery of genes and molecules at an unprecedented pace. Finally, we discuss the new challenges faced by the field of natural products and the future of systematic heterologous expression and strain-independent discovery, which promises to deliver more molecules in vials than ever before.

Principle and application of self-transcribing active regulatory region sequencing in enhancer discovery research.

Self-transcribing active regulatory region sequencing (STARR-seq) is a high-throughput sequencing method capable of simultaneously discovering and validating all enhancers within the genome. In this method, candidate sequences are inserted into plasmid vectors and electroporated into cells. Acting as both enhancers and target genes, the self-transcription of these sequences will also be enhanced by themselves. By sequencing the transcriptome and comparing the results with the non-inserted control, the locations and activity of enhancers can be determined. In traditional enhancer discovery strategies, the chromatin open regions and transcription active regions were sequenced and predicted as enhancers. However, the activity of these putative enhancers could only be validated one by one without a high-throughput method. STARR-seq solved this limitation, allowing simultaneous enhancers discovery and activity validation in a high-throughput manner. Since the introduction of STARR-seq, it has been widely used to discover enhancers and validate enhancer activity in a number of organisms and cells. In this review, we present the traditional enhancer prediction methods and the basic principles, development history, specific applications of STARR-seq, and its future prospects, aiming to provide a reference for researchers in related fields conducting enhancer studies.

Polygenic scores and Mendelian randomization identify plasma proteins causally implicated in Alzheimer's disease.

An increasing body of evidence suggests that neuroinflammation is one of the key drivers of late-onset Alzheimer's disease (LOAD) pathology. Due to the increased permeability of the blood-brain barrier (BBB) in older adults, peripheral plasma proteins can infiltrate the central nervous system (CNS) and drive neuroinflammation through interactions with neurons and glial cells. Because these inflammatory factors are heritable, a greater understanding of their genetic relationship with LOAD could identify new biomarkers that contribute to LOAD pathology or offer protection against it. We used a genome-wide association study (GWAS) of 90 different plasma proteins (n = 17,747) to create polygenic scores (PGSs) in an independent discovery (cases = 1,852 and controls = 1,990) and replication (cases = 799 and controls = 778) cohort. Multivariate logistic regression was used to associate the plasma protein PGSs with LOAD diagnosis while controlling for age, sex, principal components 1-2, and the number of APOE-e4 alleles as covariates. After meta-analyzing the PGS-LOAD associations between the two cohorts, we then performed a two-sample Mendelian randomization (MR) analysis using the summary statistics of significant plasma protein level PGSs in the meta-analysis as an exposure, and a GWAS of clinically diagnosed LOAD (cases = 21,982, controls = 41,944) as an outcome to explore possible causal relationships between the two. We identified four plasma protein level PGSs that were significantly associated (FDR-adjusted p < 0.05) with LOAD in a meta-analysis of the discovery and replication cohorts: CX3CL1, hepatocyte growth factor (HGF), TIE2, and matrix metalloproteinase-3 (MMP-3). When these four plasma proteins were used as exposures in MR with LOAD liability as the outcome, plasma levels of HGF were inferred to have a negative causal relationship with the disease when single-nucleotide polymorphisms (SNPs) used as instrumental variables were not restricted to cis-variants (OR/95%CI = 0.945/0.906-0.984, p = 0.005). Our results show that plasma HGF has a negative causal relationship with LOAD liability that is driven by pleiotropic SNPs possibly involved in other pathways. These findings suggest a low transferability between PGS and MR approaches, and future research should explore ways in which LOAD and the plasma proteome may interact.

Combining Partial True Discovery Guarantee Procedures.

Closed testing has recently been shown to be optimal for simultaneous true discovery proportion control. It is, however, challenging to construct true discovery guarantee procedures in such a way that it focuses power on some feature sets chosen by users based on their specific interest or expertise. We propose a procedure that allows users to target power on prespecified feature sets, that is, "focus sets." Still, the method also allows inference for feature sets chosen post hoc, that is, "nonfocus sets," for which we deduce a true discovery lower confidence bound by interpolation. Our procedure is built from partial true discovery guarantee procedures combined with Holm's procedure and is a conservative shortcut to the closed testing procedure. A simulation study confirms that the statistical power of our method is relatively high for focus sets, at the cost of power for nonfocus sets, as desired. In addition, we investigate its power property for sets with specific structures, for example, trees and directed acyclic graphs. We also compare our method with AdaFilter in the context of replicability analysis. The application of our method is illustrated with a gene ontology analysis in gene expressiondata.

Simultaneous discovery of compounds dominated by either molding kinetics or geographical region of origin for moisture damaged cacao beans using orthogonally applied tile-based fisher ratio analysis of GC×GC-TOFMS data

Herein, two “orthogonal” characteristics of moisture damaged cacao beans (temporally dependent molding kinetics versus the time-independent geographical region of origin) are simultaneously analyzed in a comprehensive two-dimensional (2D) gas chromatography time-of-flight mass spectrometry (GC×GC-TOFMS) dataset using tile-based Fisher ratio (F-ratio) analysis. Cacao beans from six geographical regions were analyzed once a day for six days following the initiation of moisture damage to trigger the molding process. Thus, there are two “extremes” to the experimental sample class design: six time points for the molding kinetics versus the six geographical regions of origin, resulting in a 6 × 6 element signal array referred to as a composite chemical fingerprint (CCF) for each analyte. Usually, this study would involve initial generation of two separate hit lists using F-ratio analysis, one hit list from inputting the data with the six time point classes, then another hit list from inputting the dataset from the perspective of geographic region of origin. However, analysis of two separate hit lists with the intent to distill them down to one hit list is extremely time-consuming and fraught with shortcomings due to the challenges associated with attempting to match analytes across two hit lists. To address this challenge, tile-based F-ratio analysis is “orthogonally applied” to each analyte CCF to simultaneously determine two F-ratios at the chromatographic 2D location (F-ratiokinetic and F-ratioregion) for each hit, by ranking a single hit list using the higher of the two F-ratios resulting in the discovery of 591 analytes. Further, using a pseudo-null distribution approach, at the 99.9% threshold over 400 analytes were deemed suitable for PCA classification. Using a more stringent 99.999% threshold, over 100 analytes were explored more deeply using PARAFAC to provide a purified mass spectrum.

An Independent Discovery of the 1946 Eruption of the Recurrent Nova T Coronae Borealis and an Early Prediction of a Future Outburst

One of the earliest independent discoveries of the 1946 eruption of the recurrent nova T CrB was by British amateur variable star observer, N.F.H. Knight, on 1946 February 9 at 05.40 UT. Correspondence between Knight and W.M. Lindley, the Director of the British Astronomical Association Variable Star Section, reveals what might be the first prediction of a future eruption in 2026–27.

Model Pembelajaran Self Organized Learning Environment (SOLE) Terhadap Berpikir Kritis Peserta Didik Kelas V Sekolah Dasar

This research aims to investigate the effect of implementing the Self-Organized Learning Environment (SOLE) learning model on the critical thinking abilities of class V students at SD Negeri 1 Harapan Jaya, Bandar Lampung. This learning model is known as an innovative approach that encourages independent, collaborative learning and scientific discovery The research method used was quasi-experimental with a posttest control group research design. This research sample consisted of two classes, namely the control class which received conventional learning and the experimental class which received learning using the Self Organized Learning Environment (SOLE) model. Data was collected through previously validated critical thinking test instruments. The research results showed that there was a significant increase in the critical thinking abilities of students who received learning through the Self Organized Learning Environment (SOLE) model compared to the control class. These findings indicate that the application of the Self Organized Learning Environment (SOLE) Model is able to encourage the development of critical thinking skills in students at the elementary school level. Apart from that, the results of this research also provide an implication of this research regarding the importance of implementing innovative learning models such as SOLE in improving students' critical thinking skills at the elementary school level. Educators can integrate this model into the curriculum to provide more meaningful and in-depth learning experiences, which can help students develop their critical thinking skills from an early age.

From Lecture Theaters to Online Classrooms: Examining The Growth of the Flipped Classroom

Traditional models of education involve teachers imparting knowledge to students in a classroom setting. However, new learning theories suggest that students do not all acquire information in the same way. Some learn better through independent exploration and discovery, while others thrive in social environments with hands-on instruction. This debate around student-centered versus teacher-centered pedagogy is far from resolved. One approach that is gaining attention is the "flipped classroom," which restructures learning activities so that domestic tasks typically done as homework are performed in class, with teacher guidance. Meanwhile, lectures previously given during class time are accessed autonomously by students at home, often via online video. This model arguably provides flexibility that better caters to varied student needs and learning styles. However, there are also challenges to consider when transitioning to a flipped model. This article explores the arguments on both sides, providing a balanced perspective on the benefits and limitations of traditional and flipped approaches based on current research in educational psychology. First, the article delves into the term "flipped classroom" and its relationship with the Bloom taxonomy. It then examines different flipped classroom models and finally the benefits and limitations of the flipped classroom application.

Immune features are associated with response to neoadjuvant chemo-immunotherapy for muscle-invasive bladder cancer

Neoadjuvant cisplatin-based chemotherapy is standard of care for muscle-invasive bladder cancer (MIBC). Immune checkpoint inhibition (ICI) alone, and ICI in combination with chemotherapy, have demonstrated promising pathologic response (<pT2) in the neoadjuvant setting. In LCCC1520 (NCT02690558), a phase 2 single-arm trial of neoadjuvant chemo-immunotherapy (gemcitabine and cisplatin plus pembrolizumab; NAC-ICI) for MIBC, 22/39 patients responded (pathologic downstaging as primary outcome), as previously described. Here, we report post-hoc correlative analyses. Treatment was associated with changes in tumor mutational profile, immune gene signatures, and RNA subtype switching. Clinical response was associated with an increase in plasma IL-9 from pre-treatment to initiation of cycle 2 of therapy. Tumors harbored diverse predicted antigen landscapes that change across treatment and are associated with APOBEC, tobacco, and other etiologies. Higher pre-treatment tumor PD-L1 and TIGIT RNA expression were associated with complete response. IL-8 signature and Stroma-rich subtype were associated with improved response to NAC-ICI versus neoadjuvant ICI (ABACUS trial, NCT02662309). Plasma IL-9 represents a potential predictive biomarker of NAC-ICI response, while tumor IL-8 signature and stroma-rich subtype represent potential predictive biomarkers of response benefit of NAC-ICI over neoadjuvant ICI. Future efforts must include additional independent biomarker discovery and validation, ultimately to improve the selection of patients for ICI-related treatments.