Explore the vast range of titles available.

Feature selection is one of the main techniques used to prevent overfitting in machine learning applications. The most straightforward approach for feature selection is an exhaustive search: one can go over all possible feature combinations and pick up the model with the highest accuracy. This method together with its optimizations were actively used in biomedical research, however, publicly available implementation is missing. We present ExhauFS—the user-friendly command-line implementation of the exhaustive search approach for classification and survival regression. Aside from tool description, we included three application examples in the manuscript to comprehensively review the implemented functionality. First, we executed ExhauFS on a toy cervical cancer dataset to illustrate basic concepts. Then, multi-cohort microarray breast cancer datasets were used to construct gene signatures for 5-year recurrence classification. The vast majority of signatures constructed by ExhauFS passed 0.65 threshold of sensitivity and specificity on all datasets, including the validation one. Moreover, a number of gene signatures demonstrated reliable performance on independent RNA-seq dataset without any coefficient re-tuning, i.e. , turned out to be cross-platform. Finally, Cox survival regression models were used to fit isomiR signatures for overall survival prediction for patients with colorectal cancer. Similarly to the previous example, the major part of models passed the pre-defined concordance index threshold 0.65 on all datasets. In both real-world scenarios (breast and colorectal cancer datasets), ExhauFS was benchmarked against state-of-the-art feature selection models, including L 1 -regularized sparse models. In case of breast cancer, we were unable to construct reliable cross-platform classifiers using alternative feature selection approaches. In case of colorectal cancer not a single model passed the same 0.65 threshold. Source codes and documentation of ExhauFS are available on GitHub: https://github.com/s-a-nersisyan/ExhauFS .

Hematogenous metastasis limits the survival of colorectal cancer (CRC) patients. Here, we illuminated the roles of CD44 isoforms in this process. Isoforms 3 and 4 were predominantly expressed in CRC patients. CD44 isoform 4 indicated poor outcome and correlated with epithelial–mesenchymal transition (EMT) and decreased oxidative phosphorylation (OxPhos) in patients; opposite associations were found for isoform 3. Pan‐CD44 knockdown (kd) independently impaired primary tumor formation and abrogated distant metastasis in CRC xenografts. The xenograft tumors mainly expressed the clinically relevant CD44 isoforms 3 and 4. Both isoforms were enhanced in the paranecrotic, hypoxic tumor regions but were generally absent in lung metastases. Upon CD44 kd, tumor angiogenesis was increased in the paranecrotic areas, accompanied by reduced hypoxia‐inducible factor‐1α and CEACAM5 but increased E‐cadherin expression. Mitochondrial genes and proteins were induced upon pan‐CD44 kd, as were OxPhos genes. Hypoxia increased VEGF release from tumor spheres, particularly upon CD44 kd. Genes affected upon CD44 kd in xenografts specifically overlapped concordantly with genes correlating with CD44 isoform 4 (but not isoform 3) in patients, validating the clinical relevance of the used model and highlighting the metastasis‐promoting role of CD44 isoform 4. Pan‐CD44 knockdown decreases spontaneous metastasis in human colorectal cancer xenograft models. Concurrent intratumoral gene expression alterations significantly correlate with genes differentially regulated among CD44 isoform 4 (but not isoform 3) high vs. low patients (TCGA). The corresponding gene sets and pathways include epithelial–mesenchymal transition, angiogenesis, and OxPhos. CD44 isoform 4 (but not isoform 3) correlates with poor patient outcomes.

Colorectal cancer (CRC) is one of the most common and lethal types of cancer. Although researchers have made significant efforts to study the mechanisms underlying CRC drug resistance, our knowledge of this disease is still limited, and novel therapies are in high demand. It is urgent to find new targeted therapy considering limited chemotherapy options. KRAS mutations are the most frequent molecular alterations in CRC. However, there are no approved K-Ras targeted therapies for these tumors yet. GSK-3β is demonstrated to be a critically important kinase for the survival and proliferation of K-Ras–dependent pancreatic cancer cells. In this study, we tested combinations of standard-of-care therapy and 9-ING-41, a small molecule inhibitor of GSK-3β, in CRC cell lines and patient-derived tumor organoid models of CRC. We demonstrate that 9-ING-41 inhibits the growth of CRC cells via a distinct from chemotherapy mechanism of action. Although molecular biomarkers of 9-ING-41 efficacy are yet to be identified, the addition of 9-ING-41 to the standard-of-care drugs 5-FU and oxaliplatin could significantly enhance growth inhibition in certain CRC cells. The results of the transcriptomic analysis support our findings of cell cycle arrest and DNA repair deficiency in 9-ING-41–treated CRC cells. Notably, we find substantial similarity in the changes of the transcriptomic profile after inhibition of GSK-3β and suppression of STK33, another critically important kinase for K-Ras–dependent cells, which could be an interesting point for future research. Overall, the results of this study provide a rationale for the further investigation of GSK-3 inhibitors in combination with standard-of-care treatment of CRC.

Motivation: Feature selection is one of the main techniques used to prevent overfitting in machine learning applications. The most straightforward approach for feature selection is exhaustive search: one can go over all possible feature combinations and pick up the model with the highest accuracy. This method together with its optimizations were actively used in biomedical research, however, publicly available implementation is missing. Results: We present ExhauFS - the user-friendly command-line implementation of the exhaustive search approach for classification and survival regression. Aside from tool description, we included three application examples in the manuscript to comprehensively review the implemented function-ality. First, we executed ExhauFS on a toy cervical cancer dataset to illustrate basic concepts. Then, a multi-cohort microarray and RNA-seq breast cancer datasets were used to construct gene signatures for 5-year recurrence classification. Finally, Cox survival regression models were used to fit isomiR signatures for overall survival prediction for patients with colorectal cancer. Availability: Source codes and documentation of ExhauFS are available on GitHub: https://github.com/s-a-nersisyan/ExhauFS. Competing Interest Statement The authors have declared no competing interest.

Persons with end stage kidney disease (ESKD) on dialysis are at high risk for severe COVID-19 disease. In September 2023, 2023–2024 COVID-19 vaccination was recommended in the United States for all persons aged ≥6 months. Due to possible immune dysfunction, advanced age, and high prevalence of additional underlying conditions, including immunocompromising conditions, among individuals with ESKD, reduced vaccine effectiveness (VE) is a concern. Understanding effectiveness of 2023–2024 COVID-19 vaccine among persons with ESKD can inform COVID-19 vaccine recommendations for this population. A retrospective cohort investigation was conducted among Medicare fee-for-service beneficiaries aged ≥18 years with ESKD receiving dialysis using Medicare enrollment and claims records. Follow-up began on September 17, 2023, and continued until the earliest occurrence of claim for a COVID-19–associated outcome, other censoring event, or end of follow-up. A marginal structural Cox model was used to estimate VE (calculated as [1 – hazard ratio]*100 %), interpreted as the benefit of 2023–2024 COVID-19 vaccination compared with no 2023–2024 vaccine dose. VE was estimated by presence of additional immunocompromising conditions, age group, and time since vaccination. During September 17, 2023 – April 13, 2024, 17,749/112,250 (16 %) Medicare beneficiaries aged ≥18 years with ESKD without additional immunocompromising conditions received a 2023–2024 COVID-19 vaccine dose, with a maximum 209 days of follow-up since vaccination. During the follow-up period 6539 medically attended COVID-19 events, including 3605 COVID-19-associated hospitalizations, 789 COVID-19-associated deaths, and 896 COVID-19-associated thromboembolic events, were recorded. VE against COVID-19-associated hospitalization was 55 % (95 % confidence interval [CI]: 42 % - 65 %) at 7–59 days after vaccination and 47 % (95 % CI: 35 % – 57 %) at ≥60 days after vaccination. VE against COVID-19-associated death was 71 % (95 % CI: 46 % - 84 %) at 7–59 days after vaccination and 51 % (95 % CI: 24 % – 69 %) ≥60 days after vaccination. VE against COVID-19-associated thromboembolic events was 44 % (95 % CI, 24 %, 59 %). The 2023–2024 COVID-19 vaccines provided protection against COVID-19-associated hospitalization, death, and thromboembolic events among adults with ESKD. These data support the recommendation that adults with ESKD receive the updated COVID-19 vaccine. •End stage kidney disease (ESKD) is associated with increased risk of severe COVID-19.•Uptake of 2023–2024 COVID-19 vaccination among adults with ESKD was low.•COVID-19 vaccination was effective against severe COVID-19 among adults with ESKD.•COVD-19 vaccine effectiveness among adults with ESKD waned with more time since vaccination.