Activity of distinct growth factor receptor network components in breast tumors uncovers two biologically relevant subtypes

pyComBat, a Python tool for batch effects correction in high-throughput molecular data using empirical Bayes methods

BACKGROUND

Variability in datasets is not only the product of biological processes: they are also the product of technical biases. ComBat and ComBat-Seq are among the most widely used tools for correcting those technical biases, called batch effects, in, respectively, microarray and RNA-Seq expression data.

RESULTS

In this technical note, we present a new Python implementation of ComBat and ComBat-Seq. While the mathematical framework is strictly the same, we show here that our implementations: (i) have similar results in terms of batch effects correction; (ii) are as fast or faster than the original implementations in R and; (iii) offer new tools for the bioinformatics community to participate in its development. pyComBat is implemented in the Python language and is distributed under GPL-3.0 ( https://www.gnu.org/licenses/gpl-3.0.en.html ) license as a module of the inmoose package. Source code is available at https://github.com/epigenelabs/inmoose and Python package at https://pypi.org/project/inmoose .

CONCLUSIONS

We present a new Python implementation of state-of-the-art tools ComBat and ComBat-Seq for the correction of batch effects in microarray and RNA-Seq data. This new implementation, based on the same mathematical frameworks as ComBat and ComBat-Seq, offers similar power for batch effect correction, at reduced computational cost.

Comparative RNA-sequencing analysis of the prostate in a mouse model of benign prostatic hyperplasia with bladder outlet obstruction

In ageing men, benign prostatic hyperplasia (BPH) is a chronic disease that leads to progressive lower urinary tract symptoms (LUTS) caused by obstruction of the bladder outlet (BOO). Patients with LUTS (such as increased frequency and urgency of urination) and complications of BOO (such as hydronephrosis and bladder stones) are at risk of serious health problems. BPH causes a rapidly rising burden of LUTS far exceeding that of other urological conditions. Treatment outcomes are unsatisfactory for BPH largely due to the lacking of fully understanding of the pathogenesis. Hormonal imbalances related to androgen and oestrogen can cause BPH, but the exact mechanism is still unknown, even the animal model is not fully understood. Additionally, there are no large-scale data to explain this mechanism. A BPH mouse model was established using mixed slow-release pellets of testosterone (T) and estradiol (E2), and we measured gene expression in mouse prostate tissue using RNA-seq, verified the results using qRT‒PCR, and used bioinformatics methods to analyse the differentially expressed genes (DEGs).

Fusion protein-driven IGF-IR/PI3K/AKT signals deregulate Hippo pathway promoting oncogenic cooperation of YAP1 and FUS-DDIT3 in myxoid liposarcoma

Myxoid liposarcoma (MLS) represents a common subtype of liposarcoma molecularly characterized by a recurrent chromosomal translocation that generates a chimeric FUS-DDIT3 fusion gene. The FUS-DDIT3 oncoprotein has been shown to be crucial in MLS pathogenesis. Acting as a transcriptional dysregulator, FUS-DDIT3 stimulates proliferation and interferes with adipogenic differentiation. As the fusion protein represents a therapeutically challenging target, a profound understanding of MLS biology is elementary to uncover FUS-DDIT3-dependent molecular vulnerabilities. Recently, a specific reliance on the Hippo pathway effector and transcriptional co-regulator YAP1 was detected in MLS; however, details on the molecular mechanism of FUS-DDIT3-dependent YAP1 activation, and YAP1´s precise mode of action remain unclear. In elaborate in vitro studies, employing RNA interference-based approaches, small-molecule inhibitors, and stimulation experiments with IGF-II, we show that FUS-DDIT3-driven IGF-IR/PI3K/AKT signaling promotes stability and nuclear accumulation of YAP1 via deregulation of the Hippo pathway. Co-immunoprecipitation and proximity ligation assays revealed nuclear co-localization of FUS-DDIT3 and YAP1/TEAD in FUS-DDIT3-expressing mesenchymal stem cells and MLS cell lines. Transcriptome sequencing of MLS cells demonstrated that FUS-DDIT3 and YAP1 co-regulate oncogenic gene signatures related to proliferation, cell cycle progression, apoptosis, and adipogenesis. In adipogenic differentiation assays, we show that YAP1 critically contributes to FUS-DDIT3-mediated adipogenic differentiation arrest. Taken together, our study provides mechanistic insights into a complex FUS-DDIT3-driven network involving IGF-IR/PI3K/AKT signals acting on Hippo/YAP1, and uncovers substantial cooperative effects of YAP1 and FUS-DDIT3 in the pathogenesis of MLS.

POIBM: Batch correction of heterogeneous RNA-seq datasets through latent sample matching

MOTIVATION

RNA sequencing and other high-throughput technologies are essential in understanding complex diseases, such as cancers, but are susceptible to technical factors manifesting as patterns in the measurements. These batch patterns hinder the discovery of biologically relevant patterns. Unbiased batch effect correction in heterogeneous populations currently requires special experimental designs or phenotypic labels, which are not readily available for patient samples in existing datasets.

RESULTS

We present POIBM, an RNA-seq batch correction method, which learns virtual reference samples directly from the data. We use a breast cancer cell line dataset to show that POIBM exceeds or matches the performance of previous methods, while being blind to the phenotypes. Further, we analyze The Cancer Genome Atlas RNA-seq data to show that batch effects plague many cancer types; POIBM effectively discovers the true replicates in stomach adenocarcinoma; and integrating the corrected data in endometrial carcinoma improves cancer subtyping.

AVAILABILITY

https://bitbucket.org/anthakki/poibm/ (archived at https://doi.org/10.5281/zenodo.6122436).

Tepoxalin increases chemotherapy efficacy in drug-resistant breast cancer cells overexpressing the multidrug transporter gene ABCB1

•

The COX-2 encoding gene PTGS2 is up-regulated upon ABCB1 overexpression in mammary epithelial cells.

•

The 5-LOX, COX-1/2 inhibitor tepoxalin plus chemotherapy improves treatment efficacy in ABCB1-expressing cells.

•

Tepoxalin reduces chemotherapy-induced selection for drug-resistant ABCB1-expressing cells.

Effective cancer chemotherapy treatment requires both therapy delivery and retention by malignant cells. Cancer cell overexpression of the multidrug transmembrane transporter gene ABCB1 (MDR1, multi-drug resistance protein 1) thwarts therapy retention, leading to a drug-resistant phenotype. We explored the phenotypic impact of ABCB1 overexpression in normal human mammary epithelial cells (HMECs) via acute adenoviral delivery and in breast cancer cell lines with stable integration of inducible ABCB1 expression. One hundred sixty-two genes were differentially expressed between ABCB1-expressing and GFP-expressing HMECs, including the gene encoding the cyclooxygenase-2 protein, PTGS2. Several breast cancer cell lines with inducible ABCB1 expression demonstrated sensitivity to the 5-lipoxygenase, cyclooxygenase-1/2 inhibitor tepoxalin in two-dimensional drug response assays, and combination treatment of tepoxalin either with chemotherapies or with histone deacetylase (HDAC) inhibitors improved therapeutic efficacy in these lines. Moreover, selection for the ABCB1-expressing cell population was reduced in three-dimensional co-cultures of ABCB1-expressing and GFP-expressing cells when chemotherapy was given in combination with tepoxalin. Further study is warranted to ascertain the clinical potential of tepoxalin, an FDA-approved therapeutic for use in domesticated mammals, to restore chemosensitivity and improve drug response in patients with ABCB1-overexpressing drug-resistant breast cancers.

Evaluation of Somatic Mutations in Solid Metastatic Pan-Cancer Patients

Metastasis continues to be the primary cause of all cancer-related deaths despite the recent advancements in cancer treatments. To evaluate the role of mutations in overall survival (OS) and treatment outcomes, we analyzed 957 metastatic patients with seven major cancer types who had available molecular testing results with a FoundationOne CDx® panel. The most prevalent genes with somatic mutations were TP53, KRAS, APC, and LRP1B. In this analysis, these genes had mutation frequencies higher than in publicly available datasets. We identified that the somatic mutations were seven mutually exclusive gene pairs and an additional fifty-two co-occurring gene pairs. Mutations in the mutually exclusive gene pair APC and CDKN2A showed an opposite effect on the overall survival. However, patients with CDKN2A mutations showed significantly shorter OS (HR: 1.72, 95% CI: 1.34-2.21, p < 0.001) after adjusting for cancer type, age at diagnosis, and sex. Five-year post metastatic diagnosis survival analysis showed a significant improvement in OS (median survival 28 and 43 months in pre-2015 and post-2015 metastatic diagnosis, respectively, p = 0.00021) based on the year of metastatic diagnosis. Although the use of targeted therapies after metastatic diagnosis prolonged OS, the benefit was not statistically significant. However, longer five-year progression-free survival (PFS) was significantly associated with targeted therapy use (median 10.9 months (CI: 9.7-11.9 months) compared to 9.1 months (CI: 8.1-10.1 months) for non-targeted therapy, respectively, p = 0.0029). Our results provide a clinically relevant overview of the complex molecular landscape and survival mechanisms in metastatic solid cancers.

Biological significance of KRAS mutant allele expression in ovarian endometriosis

KRAS is the most frequently mutated in ovarian endometriosis. However, it is unclear whether the KRAS mutant allele's mRNA is expressed and plays a biological role in ovarian endometriosis. Here, we performed mutation-specific RNA in situ hybridization to evaluate mutant allele expression of KRAS p.G12V, the most frequently detected mutation in ovarian endometriosis in our previous study, in formalin-fixed paraffin-embedded tissue (FFPE) samples of ovarian endometriosis, cancer cell lines, and ovarian cancers. First, we verified that mutant or wild-type allele of KRAS were expressed in all 5 cancer cell lines and 9 ovarian cancer cases corresponding to the mutation status. Next, we applied this assay to 26 ovarian endometriosis cases, and observed mutant allele expression of KRAS p.G12V in 10 cases. Mutant or wild-type allele of KRAS were expressed in line with mutation status in 12 available endometriosis cases for which KRAS gene sequence was determined. Comparison of clinical features between ovarian endometriosis with KRAS p.G12V mutant allele expression and with KRAS wild-type showed that KRAS p.G12V mutant allele expression was significantly associated with inflammation in ovarian endometriosis. Finally, we assessed the spatial distribution of KRAS mutant allele expression in 5 endometriosis cases by performing multiregional sampling. Intratumor heterogeneity of KRAS mutant allele expression was observed in two endometriosis cases, whereas the spatial distribution of KRAS p.G12V mutation signals were diffuse and homogenous in ovarian cancer. In conclusion, evaluation of oncogene mutant expression will be useful for clarifying the biological significance of oncogene mutations in benign tumors.

SARS-CoV-2 early infection signature identified potential key infection mechanisms and drug targets

Background

The ongoing COVID-19 outbreak has caused devastating mortality and posed a significant threat to public health worldwide. Despite the severity of this illness and 2.3 million worldwide deaths, the disease mechanism is mostly unknown. Previous studies that characterized differential gene expression due to SARS-CoV-2 infection lacked robust validation. Although vaccines are  now available, effective treatment options are still out of reach.

Results

To characterize the transcriptional activity of SARS-CoV-2 infection, a gene signature consisting of 25 genes was generated using a publicly available RNA-Sequencing (RNA-Seq) dataset of cultured cells infected with SARS-CoV-2. The signature estimated infection level accurately in bronchoalveolar lavage fluid (BALF) cells and peripheral blood mononuclear cells (PBMCs) from healthy and infected patients (mean 0.001 vs. 0.958; P < 0.0001). These signature genes were investigated in their ability to distinguish the severity of SARS-CoV-2 infection in a single-cell RNA-Sequencing dataset. TNFAIP3, PPP1R15A, NFKBIA, and IFIT2 had shown bimodal gene expression in various immune cells from severely infected patients compared to healthy or moderate infection cases. Finally, this signature was assessed using the publicly available ConnectivityMap database to identify potential disease mechanisms and drug repurposing candidates. Pharmacological classes of tricyclic antidepressants, SRC-inhibitors, HDAC inhibitors, MEK inhibitors, and drugs such as atorvastatin, ibuprofen, and ketoconazole showed strong negative associations (connectivity score < − 90), highlighting the need for further evaluation of these candidates for their efficacy in treating SARS-CoV-2 infection.

Conclusions

Thus, using the 25-gene SARS-CoV-2 infection signature, the SARS-CoV-2 infection status was captured in BALF cells, PBMCs and postmortem lung biopsies. In addition, candidate SARS-CoV-2 therapies with known safety profiles were identified. The signature genes could potentially also be used to characterize the COVID-19 disease severity in patients’ expression profiles of BALF cells.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12864-021-07433-4.

ComBat-seq: batch effect adjustment for RNA-seq count data

The benefit of integrating batches of genomic data to increase statistical power is often hindered by batch effects, or unwanted variation in data caused by differences in technical factors across batches. It is therefore critical to effectively address batch effects in genomic data to overcome these challenges. Many existing methods for batch effects adjustment assume the data follow a continuous, bell-shaped Gaussian distribution. However in RNA-seq studies the data are typically skewed, over-dispersed counts, so this assumption is not appropriate and may lead to erroneous results. Negative binomial regression models have been used previously to better capture the properties of counts. We developed a batch correction method, ComBat-seq, using a negative binomial regression model that retains the integer nature of count data in RNA-seq studies, making the batch adjusted data compatible with common differential expression software packages that require integer counts. We show in realistic simulations that the ComBat-seq adjusted data results in better statistical power and control of false positives in differential expression compared to data adjusted by the other available methods. We further demonstrated in a real data example that ComBat-seq successfully removes batch effects and recovers the biological signal in the data.

Pathway activity profiling of growth factor receptor network and stemness pathways differentiates metaplastic breast cancer histological subtypes

Background

Gene expression profiling of rare cancers has proven challenging due to limited access to patient materials and requirement of intact, non-degraded RNA for next-generation sequencing. We customized a gene expression panel compatible with degraded RNA from formalin-fixed, paraffin-embedded (FFPE) patient cancer samples and investigated its utility in pathway activity profiling in patients with metaplastic breast cancer (MpBC).

Methods

Activity of various biological pathways was profiled in samples from nineteen patients with MpBC and 8 patients with invasive ductal carcinoma with triple negative breast cancer (TNBC) phenotype using a custom gene expression-based assay of 345 genes.

Results

MpBC samples of mesenchymal (chondroid and/or osteoid) histology demonstrated increased SNAI1 and BCL2L11 pathway activity compared to samples with non-mesenchymal histology. Additionally, late cornified envelope and keratinization genes were downregulated in MpBC compared to TNBC, and epithelial-to-mesenchymal transition (EMT) and collagen genes were upregulated in MpBC. Patients with high activity of an invasiveness gene expression signature, as well as high expression of the mesenchymal marker and extracellular matrix glycoprotein gene SPARC, experienced worse outcomes than those with low invasiveness activity and low SPARC expression.

Conclusions

This study demonstrates the utility of gene expression profiling of metaplastic breast cancer FFPE samples with a custom counts-based assay. Gene expression patterns identified by this assay suggest that, although often histologically triple negative, patients with MpBC have distinct pathway activation compared to patients with invasive ductal TNBC. Incorporation of targeted therapies may lead to improved outcome for MpBC patients, especially in those patients expressing increased activity of invasiveness pathways.

Electronic supplementary material

The online version of this article (10.1186/s12885-019-6052-z) contains supplementary material, which is available to authorized users.

Alternative empirical Bayes models for adjusting for batch effects in genomic studies

Background

Combining genomic data sets from multiple studies is advantageous to increase statistical power in studies where logistical considerations restrict sample size or require the sequential generation of data. However, significant technical heterogeneity is commonly observed across multiple batches of data that are generated from different processing or reagent batches, experimenters, protocols, or profiling platforms. These so-called batch effects often confound true biological relationships in the data, reducing the power benefits of combining multiple batches, and may even lead to spurious results in some combined studies. Therefore there is significant need for effective methods and software tools that account for batch effects in high-throughput genomic studies.

Results

Here we contribute multiple methods and software tools for improved combination and analysis of data from multiple batches. In particular, we provide batch effect solutions for cases where the severity of the batch effects is not extreme, and for cases where one high-quality batch can serve as a reference, such as the training set in a biomarker study. We illustrate our approaches and software in both simulated and real data scenarios.

Conclusions

We demonstrate the value of these new contributions compared to currently established approaches in the specified batch correction situations.

Electronic supplementary material

The online version of this article (10.1186/s12859-018-2263-6) contains supplementary material, which is available to authorized users.

Combating subclonal evolution of resistant cancer phenotypes

Metastatic breast cancer remains challenging to treat, and most patients ultimately progress on therapy. This acquired drug resistance is largely due to drug-refractory sub-populations (subclones) within heterogeneous tumors. Here, we track the genetic and phenotypic subclonal evolution of four breast cancers through years of treatment to better understand how breast cancers become drug-resistant. Recurrently appearing post-chemotherapy mutations are rare. However, bulk and single-cell RNA sequencing reveal acquisition of malignant phenotypes after treatment, including enhanced mesenchymal and growth factor signaling, which may promote drug resistance, and decreased antigen presentation and TNF-α signaling, which may enable immune system avoidance. Some of these phenotypes pre-exist in pre-treatment subclones that become dominant after chemotherapy, indicating selection for resistance phenotypes. Post-chemotherapy cancer cells are effectively treated with drugs targeting acquired phenotypes. These findings highlight cancer's ability to evolve phenotypically and suggest a phenotype-targeted treatment strategy that adapts to cancer as it evolves.