Human cells consist of a complex hierarchy of components, many of which remain unexplored 1,2 . Here we construct a global map of human subcellular architecture through joint measurement of biophysical interactions and immunofluorescence images for over 5,100 proteins in U2OS osteosarcoma cells. Self-supervised multimodal data integration resolves 275 molecular assemblies spanning the range of 10 −8 to 10 −5 m, which we validate systematically using whole-cell size-exclusion chromatography and annotate using large language models 3 . We explore key applications in structural biology, yielding structures for 111 heterodimeric complexes and an expanded Rag–Ragulator assembly. The map assigns unexpected functions to 975 proteins, including roles for C18orf21 in RNA processing and DPP9 in interferon signalling, and identifies assemblies with multiple localizations or cell type specificity. It decodes paediatric cancer genomes 4 , identifying 21 recurrently mutated assemblies and implicating 102 validated new cancer proteins. The associated Cell Visualization Portal and Mapping Toolkit provide a reference platform for structural and functional cell biology.

Background Surgical drains are a key component for recovery in breast reconstruction procedures. However, they are often cumbersome and carry a risk of infection with prolonged use. We aimed to develop a more thorough understanding of patient and health care provider perspectives on surgical drains, to inform future efforts in improving the breast reconstruction patient experience. Methods Twenty-nine breast reconstruction patients and eight plastic surgery providers were recruited to complete surveys focused on surgical drains. Likert scales ranging from 1 to 5 were developed to gauge how bothersome drains felt, as well as concern for infection. Ordinal variable and categorical multiple-choice analyses were applied as appropriate. Results Fifteen (51.7%) patients underwent implant-based breast reconstruction, and 14 (48.3%) patients underwent autologous breast reconstruction. The most common duration of drain placement was 2 weeks (N = 13). The surgical site infection (SSI) rate requiring antibiotics was 28% (N = 8). On a scale of 1 to 5, both patients (median = 3) and providers (median = 2.5) viewed drains as bothersome. Patients were “frequently” concerned about infection risk (median = 3). Other high-frequency patient concerns included general pain and discomfort. Conclusion Surgical drains are a common component of breast reconstruction procedures and are viewed as cumbersome by both patients and providers. Patients expressed concerns about drain site pain, discomfort, and tugging on clothing. Patients and providers both believed that drains could contribute to SSI. Overall, these data provide insight to drive future improvements in the patient drain experience.

Supplementary Figure S1. Aspects of replication stress addressed by cancer therapeutics. Supplementary Figure S2. Nested systems in tumors (NeST). Supplementary Figure S3. VNN schematic. Supplementary Figure S4. Performance and interpretation of the multi-drug VNN. Supplementary Figure S5. Evaluation of assemblies by systematic drug sensitivity screens. Supplementary Figure S6. Survival analysis of cisplatin-treated TCGA cohorts.

Gene set analysis is a mainstay of functional genomics, but it relies on manually curated databases of gene functions that are incomplete and unaware of biological context. Here we evaluate the ability of OpenAI's GPT-4, a Large Language Model (LLM), to develop hypotheses about common gene functions from its embedded biomedical knowledge. We created a GPT-4 pipeline to label gene sets with names that summarize their consensus functions, substantiated by analysis text and citations. Benchmarking against named gene sets in the Gene Ontology, GPT-4 generated very similar names in 50% of cases, while in most remaining cases it recovered the name of a more general concept. In gene sets discovered in 'omics data, GPT-4 names were more informative than gene set enrichment, with supporting statements and citations that largely verified in human review. The ability to rapidly synthesize common gene functions positions LLMs as valuable functional genomics assistants.

Abstract Rapid proliferation is a hallmark of cancer associated with sensitivity to therapeutics that cause DNA replication stress (RS). Many tumors exhibit drug resistance, however, via molecular pathways that are incompletely understood. Here, we develop an ensemble of predictive models that elucidate how cancer mutations impact the response to common RS-inducing (RSi) agents. The models implement recent advances in deep learning to facilitate multidrug prediction and mechanistic interpretation. Initial studies in tumor cells identify 41 molecular assemblies that integrate alterations in hundreds of genes for accurate drug response prediction. These cover roles in transcription, repair, cell-cycle checkpoints, and growth signaling, of which 30 are shown by loss-of-function genetic screens to regulate drug sensitivity or replication restart. The model translates to cisplatin-treated cervical cancer patients, highlighting an RTK–JAK–STAT assembly governing resistance. This study defines a compendium of mechanisms by which mutations affect therapeutic responses, with implications for precision medicine. Significance: Zhao and colleagues use recent advances in machine learning to study the effects of tumor mutations on the response to common therapeutics that cause RS. The resulting predictive models integrate numerous genetic alterations distributed across a constellation of molecular assemblies, facilitating a quantitative and interpretable assessment of drug response. This article is featured in Selected Articles from This Issue, p. 384

Surgical drains are a critical component of the recovery process for breast reconstruction procedures, reducing the incidence of seroma and excess fluid collection. 1 Scomacao I. Cummins A. Roan E. Duraes E.F. Djohan R. The use of surgical site drains in breast reconstruction: a systematic review. J Plast Reconstr Aesthetic Surg. 2020; 73: 651-662 Abstract Full Text Full Text PDF PubMed Scopus (14) Google Scholar As such, drains are a standard of care practice used by a majority of plastic surgeons in breast reconstruction. 2 Chua C. Bascone C.M. Pereira C. et al. Final 24-hour drain output and postoperative day are poor indicators for appropriate drain removal. Plast Reconstr Surg Glob Open. 2022; 10: e4160 Crossref PubMed Scopus (1) Google Scholar However, prolonged drain duration is an independent risk factor for surgical site infection (SSI), 3 Hanna K.R. Tilt A. Holland M. et al. Reducing infectious complications in implant based breast reconstruction: impact of early expansion and prolonged drain use. Ann Plast Surg. 2016; 76: S312-S315 Crossref PubMed Scopus (44) Google Scholar and there are currently no widely accepted methods for cleaning surgical drains.

Abstract Pediatric tumors are characterized by a low incidence of somatic mutations, which occur in different genomic patterns from adult cancers. For these reasons, interpreting the functional impact of mutations in pediatric cancer patients has remained difficult. One approach to interpreting rare coding mutations is to analyze the convergence of these mutations on larger structural and functional units, such as protein complexes, super assemblies, biomolecular condensates, and organelles. Towards this aim, we created a cancer cell map in the osteosarcoma cell line based on proteome-wide integration of protein interactions and immunofluorescent imaging in U-2 OS pediatric bone cancer cells. This multiscale integrated cell map of U-2 OS cell organizes 5,254 proteins into 270 protein assemblies. Using this multiscale cell map, we analyze 772 genomes from 18 cancer types reported in the DKFZ pan-pediatric cancer study to identify 25 protein assemblies in the map under mutational pressure. This analysis reveals mutated protein assemblies involved with chromatin maintenance and remodeling as well as cell integrity and mobility. For example, we identify a Collagen Biosynthesis and Remodeling Complex, as mutated in 9 cancer types and 15 patients in the dataset, indicating extracellular matrix remodeling is associated with pediatric cancer progression. We also identify a putative biological condensate, which we call the Nuclear Transcription Suppression Complex (NTSC). This protein assembly is found mutated in 20 patients across 10 cancer types, implicating intrinsically disordered proteins in development of pediatric cancer. Overall, this work illustrates how a global multiscale map of cell architecture enables a mechanistic understanding of clinical patient genomes. Citation Format: Mengzhou Hu, Leah V. Schaffer, Edward L. Huttlin, Gege Qian, Andrew Latham, Abantika Pal, Laura Pontano Vaites, Trang Le, Yue Qin, Christopher Churas, Dexter Pratt, Robin Bachelder, Peter Zage, Ignacia Echeverria, Andrej Šali, J. Wade Harper, Steven P. Gygi, Emma Lundberg, Trey Ideker. A map of osteosarcoma cell architecture reveals convergence of pediatric cancer mutations on protein assemblies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6543.

Abstract Rapid proliferation is a hallmark of tumor cells, associated with sensitivity to therapeutics that cause DNA replication stress (RS). Many tumors exhibit drug resistance, however, via molecular pathways that are incompletely understood. Here, we develop an ensemble of predictive models that elucidate how cancer mutations impact the response to common RS-inducing (RSi) agents. The models implement recent advances in deep learning to facilitate multi-drug prediction and mechanistic interpretation. Instead of associating individual genetic alterations with RSi drug responses directly, the strategy is to project these alterations on a map of protein complexes and larger molecular assemblies associated with cancer. Initial studies in tumor cells identify 41 molecular assemblies that integrate alterations in hundreds of genes for accurate drug response prediction. These cover roles in transcription, repair, cell-cycle checkpoints, and growth signaling, of which 30 are shown by loss-of-function genetic screens to regulate drug sensitivity or replication restart. The model translates to cisplatin-treated cervical cancer patients, highlighting an RTK-JAK-STAT assembly governing resistance. This study defines a compendium of mechanisms by which mutations affect therapeutic responses, with implications for drug selection and combination. Citation Format: Xiaoyu Zhao, Akshat Singhal, SungJoon Park, JungHo Kong, Robin Bachelder, Trey Ideker. Cancer mutations converge on a constellation of molecular assemblies to predict resistance to replication stress agents [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4918.

Table S1 Annotation of NeST assembly.xlsx

Cyclin-dependent kinase 4 and 6 inhibitors (CDK4/6is) have revolutionized breast cancer therapy. However, <50% of patients have an objective response, and nearly all patients develop resistance during therapy. To elucidate the underlying mechanisms, we constructed an interpretable deep learning model of the response to palbociclib, a CDK4/6i, based on a reference map of multiprotein assemblies in cancer. The model identifies eight core assemblies that integrate rare and common alterations across 90 genes to stratify palbociclib-sensitive versus palbociclib-resistant cell lines. Predictions translate to patients and patient-derived xenografts, whereas single-gene biomarkers do not. Most predictive assemblies can be shown by CRISPR-Cas9 genetic disruption to regulate the CDK4/6i response. Validated assemblies relate to cell-cycle control, growth factor signaling and a histone regulatory complex that we show promotes S-phase entry through the activation of the histone modifiers KAT6A and TBL1XR1 and the transcription factor RUNX1. This study enables an integrated assessment of how a tumor's genetic profile modulates CDK4/6i resistance.

While immune checkpoint inhibitors have revolutionized cancer therapy, many patients exhibit poor outcomes. Here, we show immunotherapy responses in bladder and non-small cell lung cancers are effectively predicted by factoring tumor mutation burden (TMB) into burdens on specific protein assemblies. This approach identifies 13 protein assemblies for which the assembly-level mutation burden (AMB) predicts treatment outcomes, which can be combined to powerfully separate responders from nonresponders in multiple cohorts (e.g., 76% versus 37% bladder cancer 1-year survival). These results are corroborated by (i) engineered disruptions in the predictive assemblies, which modulate immunotherapy response in mice, and (ii) histochemistry showing that predicted responders have elevated inflammation. The 13 assemblies have diverse roles in DNA damage checkpoints, oxidative stress, or Janus kinase/signal transducers and activators of transcription signaling and include unexpected genes (e.g., PIK3CG and FOXP1) for which mutation affects treatment response. This study provides a roadmap for using tumor cell biology to factor mutational effects on immune response.

Table S2 Odds ratio for 45 drug response predictions with multi-drug model.xlsx

Table S4 Assembly enrichment in replication and drug sensitivity screens.xlsx

Table S5 Assembly predictive performance in cisplatin-treated cervical cancer.xlsx

Table S3 Importance of assemblies in VNN models.xlsx

Ascites from ovarian cancer-bearing mice contributes to a faster progression of the cancer.

MemGRP78 positive tumor cells have a lower proliferation rate

Characterization of ID8 cell line

MemGRP78 positive tumor cells have a lower proliferation rate