Building gene expression profile classifiers with a simple and efficient rejection option in R

Elucidating prognosis in cervical squamous cell carcinoma and endocervical adenocarcinoma: a novel anoikis-related gene signature model

Background

Cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC) are among the most prevalent gynecologic malignancies globally. The prognosis is abysmal once cervical cancer progresses to lymphatic metastasis. Anoikis, a specialized form of apoptosis induced by loss of cell adhesion to the extracellular matrix, plays a critical role. The prediction model based on anoikis-related genes (ARGs) expression and clinical data could greatly aid clinical decision-making. However, the relationship between ARGs and CESC remains unclear.

Methods

ARGs curated from the GeneCards and Harmonizome portals were instrumental in delineating CESC subtypes and in developing a prognostic framework for patients afflicted with this condition. We further delved into the intricacies of the immune microenvironment and pathway enrichment across the identified subtypes. Finally, our efforts culminated in the creation of an innovative nomogram that integrates ARGs. The utility of this prognostic tool was underscored by Decision Curve Analysis (DCA), which illuminate its prospective benefits in guiding clinical interventions.

Results

In our study, We discerned a set of 17 survival-pertinent, anoikis-related differentially expressed genes (DEGs) in CESC, from which nine were meticulously selected for the construction of prognostic models. The derived prognostic risk score was subsequently validated as an autonomous prognostic determinant. Through comprehensive functional analyses, we observed distinct immune profiles and drug response patterns among divergent prognostic stratifications. Further, we integrated the risk scores with the clinicopathological characteristics of CESC to develop a robust nomogram. DCA corroborated the utility of our model, demonstrating its potential to enhance patient outcomes through tailored clinical treatment strategies.

Conclusion

The predictive signature, encompassing nine pivotal genes, alongside the meticulously constructed nomogram developed in this research, furnishes clinicians with a sophisticated tool for tailoring treatment strategies to individual patients diagnosed with CESC.

Evaluating Pointwise Reliability of Machine Learning prediction

Interest in Machine Learning applications to tackle clinical and biological problems is increasing. This is driven by promising results reported in many research papers, the increasing number of AI-based software products, and by the general interest in Artificial Intelligence to solve complex problems. It is therefore of importance to improve the quality of machine learning output and add safeguards to support their adoption. In addition to regulatory and logistical strategies, a crucial aspect is to detect when a Machine Learning model is not able to generalize to new unseen instances, which may originate from a population distant to that of the training population or from an under-represented subpopulation. As a result, the prediction of the machine learning model for these instances may be often wrong, given that the model is applied outside its "reliable" space of work, leading to a decreasing trust of the final users, such as clinicians. For this reason, when a model is deployed in practice, it would be important to advise users when the model's predictions may be unreliable, especially in high-stakes applications, including those in healthcare. Yet, reliability assessment of each machine learning prediction is still poorly addressed. Here, we review approaches that can support the identification of unreliable predictions, we harmonize the notation and terminology of relevant concepts, and we highlight and extend possible interrelationships and overlap among concepts. We then demonstrate, on simulated and real data for ICU in-hospital death prediction, a possible integrative framework for the identification of reliable and unreliable predictions. To do so, our proposed approach implements two complementary principles, namely the density principle and the local fit principle. The density principle verifies that the instance we want to evaluate is similar to the training set. The local fit principle verifies that the trained model performs well on training subsets that are more similar to the instance under evaluation. Our work can contribute to consolidating work in machine learning especially in medicine.

SMAD-6, -7 and -9 are potential molecular biomarkers for the prognosis in human lung cancer

SMADs, a family of proteins that function as signal transducers and transcriptional regulators to regulate various signaling pathways, including the transforming growth factor-β signaling pathway, are similar to the mothers against decapentaplegic family of genes and the sma gene family in Caenorhabditis elegans. SMADs generate context-dependent modulation by interacting with various sequence-specific transcription factors, such as E2F4/5, c-Fos, GATA3, YY1 and SRF, which have been found to serve a key role in lung carcinoma oncogenesis and progression. However, the prognostic values of the eight SMADs in lung cancer have not been fully understood. In the present study, the expression levels and survival data of SMADs in patients with lung carcinoma from the Oncomine, Gene Expression Profiling Interactive Analysis, Kaplan-Meier plotter and cBioPortal databases were downloaded and analyzed. It was found that the mRNA expression levels of SMAD-6, -7 and -9 were decreased in lung adenocarcinoma and squamous cell carcinoma compared with that in adjacent normal tissues, while there was no significant difference in SMADs 1-5. Survival analysis revealed that not only were low transcriptional levels of SMAD-6, -7 and -9 associated with low overall survival but they also had prognostic role for progression-free survival and post-progression survival (P<0.05) in patients with lung carcinoma. In conclusion, the present study demonstrated that SMAD-6, -7 and -9 are potential biomarkers for the prognosis of patients with lung carcinoma.

Expression of cytosolic peroxiredoxins in Plasmodium berghei ookinetes is regulated by environmental factors in the mosquito bloodmeal

The Plasmodium ookinete develops over several hours in the bloodmeal of its mosquito vector where it is exposed to exogenous stresses, including cytotoxic reactive oxygen species (ROS). How the parasite adapts to these challenging conditions is not well understood. We have systematically investigated the expression of three cytosolic antioxidant proteins, thioredoxin-1 (Trx-1), peroxiredoxin-1 (TPx-1), and 1-Cys peroxiredoxin (1-Cys Prx), in developing ookinetes of the rodent parasite Plasmodium berghei under various growth conditions. Transcriptional profiling showed that tpx-1 and 1-cys prx but not trx-1 are more strongly upregulated in ookinetes developing in the mosquito bloodmeal when compared to ookinetes growing under culture conditions. Confocal immunofluorescence imaging revealed comparable expression patterns on the corresponding proteins. 1-Cys Prx in particular exhibited strong expression in mosquito-derived ookinetes but was not detectable in cultured ookinetes. Furthermore, ookinetes growing in culture upregulated tpx-1 and 1-cys prx when challenged with exogenous ROS in a dose-dependent fashion. This suggests that environmental factors in the mosquito bloodmeal induce upregulation of cytosolic antioxidant proteins in Plasmodium ookinetes. We found that in a parasite line lacking TPx-1 (TPx-1KO), expression of 1-Cys Prx occurred significantly earlier in mosquito-derived TPx-1KO ookinetes when compared to wild type (WT) ookinetes. The protein was also readily detectable in cultured TPx-1KO ookinetes, indicating that 1-Cys Prx at least in part compensates for the loss of TPx-1 in vivo. We hypothesize that this dynamic expression of the cytosolic peroxiredoxins reflects the capacity of the developing Plasmodium ookinete to rapidly adapt to the changing conditions in the mosquito bloodmeal. This would significantly increase its chances of survival, maturation and subsequent escape. Our results also emphasize that environmental conditions must be taken into account when investigating Plasmodium-mosquito interactions.

The malaria parasite Plasmodium is transmitted by Anopheles mosquitoes. Within the midgut of the insect, it is exposed to multiple environmental stresses, including cytotoxic reactive oxygen species (ROS). To avoid destruction, the parasite develops into a motile ookinete capable of leaving the midgut. Yet, ookinete development lasts over several hours and requires the parasite to adapt to an increasingly challenging environment. Here we show that ookinetes of the rodent parasite Plasmodium berghei during development in the mosquito midgut increase the expression of the protective antioxidant proteins peroxiredoxin-1 (TPx-1) and 1-Cys peroxiredoxin (1-Cys Prx). This upregulation was also inducible in cultured ookinetes by challenging them with ROS. This suggests that ookinetes actively modulate the expression of their antioxidant proteins in response to the changing conditions in the mosquito. We also found that ookinetes lacking TPx-1 (TPx-1KO) upregulated 1-Cys Prx expression significantly earlier than wild type ookinetes. This indicates that the TPx-1KO parasites compensate for the loss of TPx-1 by altering the expression pattern of the functionally related 1-Cys Prx. The observed dynamic regulation of the cytosolic antioxidant proteins may help the Plasmodium ookinete to adapt to rapidly changing environmental conditions and thus to increase the probability of survival, maturation and escape from the mosquito midgut.

Towards big data science in the decade ahead from ten years of InCoB and the 1st ISCB-Asia Joint Conference

The 2011 International Conference on Bioinformatics (InCoB) conference, which is the annual scientific conference of the Asia-Pacific Bioinformatics Network (APBioNet), is hosted by Kuala Lumpur, Malaysia, is co-organized with the first ISCB-Asia conference of the International Society for Computational Biology (ISCB). InCoB and the sequencing of the human genome are both celebrating their tenth anniversaries and InCoB’s goalposts for the next decade, implementing standards in bioinformatics and globally distributed computational networks, will be discussed and adopted at this conference. Of the 49 manuscripts (selected from 104 submissions) accepted to BMC Genomics and BMC Bioinformatics conference supplements, 24 are featured in this issue, covering software tools, genome/proteome analysis, systems biology (networks, pathways, bioimaging) and drug discovery and design.