Search

A very important function of an issue tracking system is to assign labels to issue reports, such as bug, feature, enhancement, etc., in order to categorize issues to facilitate various development activities. In practice, it is very common that an issue has multiple labels. However, current works are mainly based on single-label prediction, which are not suitable for just-in-time multi-labeling services, due to the low efficiency. Therefore, in this paper, we propose MULA, a just-in-time MUlti-LAbeling system, which learns and automatically assigns multiple labels to issue reports. We have built a dataset with 81,601 entries and 11 labels, as the first benchmark for this task, and implemented a GitHub app. To the best of our knowledge, this is the first work and tool for online multi-labeling GitHub issues based on their categories. We conduct a comprehensive empirical study, including comparisons with five commonly adopted labeling models that show the superiority of MULA, as well as an evaluation that shows high consistency between MULA’s suggestions and developers’ opinions.

The clustering technique has attracted a lot of attention as a promising strategy for parallel debugging in multi-fault scenarios, this heuristic approach (i.e., failure indexing or fault isolation) enables developers to perform multiple debugging tasks simultaneously through dividing failed test cases into several disjoint groups. When using statement ranking representation to model failures for better clustering, several factors influence clustering effectiveness, including the risk evaluation formula (REF), the number of faults (NOF), the fault type (FT), and the number of successful test cases paired with one individual failed test case (NSP1F). In this paper, we present the first comprehensive empirical study of how these four factors influence clustering effectiveness. We conduct extensive controlled experiments on 1060 faulty versions of 228 simulated faults and 141 real faults, and the results reveal that: 1) GP19 is highly competitive across all REFs, 2) clustering effectiveness decreases as NOF increases, 3) higher clustering effectiveness is easier to achieve when a program contains only predicate faults, and 4) clustering effectiveness remains when the scale of NSP1F is reduced to 20%.

This report summarises the organisation and execution of MET 2022, the 7th International Workshop on Metamorphic Testing, which took place as a virtual event in conjunction with the ICSE 2022 conference.

Question Answering (QA) is an attractive and challenging area in NLP community. With the development of QA technique, plenty of QA software has been applied in daily human life to provide convenient access of information retrieval. To investigate the performance of QA software, many benchmark datasets have been constructed to provide various test cases. However, current QA software is mainly tested in a reference-based paradigm, in which the expected outputs (labels) of test cases are mandatory to be annotated with much human effort before testing. As a result, neither the just-in-time test during usage nor the extensible test on massive unlabeled real-life data is feasible, which keeps the current testing of QA software from being flexible and sufficient. In this work, we propose a novel testing method, QAAskeR+, with five new Metamorphic Relations for QA software. QAAskeR+ does not refer to the annotated labels of test cases. Instead, based on the idea that a correct answer should imply a piece of reliable knowledge that always conforms with any other correct answer, QAAskeR+ tests QA software by inspecting its behaviors on multiple recursively asked questions that are relevant to the same or some further enriched knowledge. Experimental results show that QAAskeR+ can reveal quite a few violations that indicate actual answering issues on various mainstream QA software without using any pre-annotated labels.

Background and objectives The application of adipose derived stem cells (ADSCs) in skin repair has attracted much attention nowadays. Epidermal growth factor (EGF) participates in the progress of skin proliferation, differentiation and so forth. We aimed to explore the role of EGF in the proliferation, invasion, migration and transdifferentiation into epidermal cell phenotypes of ADSCs. Methods and results ADSCs were extracted from adipose tissues from patient. Immunophenotyping was determined by flow cytometry. Overexpressed EGF or siEGF was transfected by lentiviruses. EGF was determined by enzyme linked immunosorbent assay (ELISA) or western blot. ADSCs and HaCaT cells were co-cultured by Transwell chambers. Conditioned medium (CM) was obtained from cultured HaCaT cells and used for the culturing of ADSCs. Cell viability was tested by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Invasion rate was measured by Transwell invasion assay and migration rate by wound healing test. mRNA and protein levels were measured by qPCR and western blot respectively. The extracted cells from adipose tissues were identified as ADSCs by morphology and immunophenotyping. The expression of EGF was up or down regulated constantly in HaCaT cell line after transfection. EGF overexpression upregulated the proliferation, migration and invasion rates of ADSCs, and EGF expression regulated the expression of cytokeratin-19 (CK19) and integrin-β as well. Conclusions EGF could be served as a stimulus to promote the proliferation, migration, and invasion as well as the transdifferentiation into epidermal stem cell immunophenotyping of ADSCs. The results showed that EGF had a promising effect on the repair of skin wound.

Background Ischemia-reperfusion (IR) injury is a main cause to and the mechanism of necrosis after flap transplantation. Researches were hardly conducted on the role and possible mechanism of keratinocyte growth factor (KGF) in association with IR flap injury. Materials and methods A CoCl2-stimulated hypoxia cell model was established to investigate the effects of KGF on cell viability, apoptosis, cell cycle, and reactive oxygen species level. The experiments were performed by cell counting kit-8 and flow cytometry as required. Meanwhile, the expressions of cell cycle-related and nuclear factor E2–related factor 2 (Nrf2) signaling–related genes were determined using quantitative real-time PCR and Western blot. The right dorsolateral areas of Institute of Cancer Research mice were marked as flaps, the pedicle of which formed an IR process through clamping and loosening. Tissue morphologies were observed using hematoxylin and eosin staining 24 h after the surgery. The effects of KGF on cell apoptosis and associated genes expressions were studied by terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling, immunohistochemistry, and Western blot. Results HaCAT cells treated with 40 μM CoCl2 could not only reduce cell viability, promote cell apoptosis, arrest G1 phase of cell cycle and increase the activity of reactive oxygen species but also downregulate the expressions of c-myc, c-fos, transforming growth factor-α, Nrf2, heme oxygenase-1, and gamma-glutamyl cysteine synthetase. Additional recombinant human KGF, on one hand, could protect the cells from hypoxia injury. On the other hand, recombinant human KGF could significantly inhibit cell apoptosis, increase KGF activity, and increase the Nrf2, heme oxygenase-1, and gamma-glutamyl cysteine synthetase proteins levels in IR flap tissues. Conclusions KGF played an important role in protecting mice flaps from IR injury, and the possible mechanism was involved in activating the Nrf2 signaling.

Deep learning has developed rapidly in recent years, attracting the attention of numerous researchers. Since a wide range of topics are covered in this field, we are wondering what topics researchers have concerned about. However, after investigation, we find that very few researchers have paid attention to this demand. In this paper, we conduct a large-scale study to analyze the questions faced by deep learning developers. We use Stack Overflow, one of the largest questiona (2) Gradient Propagation is the most popular among all the topics and (3) Object Detection is the most difficult; (4) issues of Package Installation, Code Understanding and Method Introduction are common in the current different deep learning frameworks; (5) there are three trends in these topics, e.g., a significant rising trend is found in the number of discussion on Data Shape. Finally, based on our research findings, we make some targeted and valuable suggestions for developers, researchers, educators, and framework providers of deep learning.

Machine Reading Comprehension (MRC) in Natural Language Processing has seen great progress recently. But almost all the current MRC software is validated with a reference-based method, which requires well-annotated labels for test cases and tests the software by checking the consistency between the labels and the outputs. However, labeling test cases of MRC could be very costly due to their complexity, which makes reference-based validation hard to be extensible and sufficient. Furthermore, solely checking the consistency and measuring the overall score may not be sensible and flexible for assessing the language understanding capability. In this paper, we propose a property-based validation method for MRC software with Metamorphic Testing to supplement the reference-based validation. It does not refer to the labels and hence can make much data available for testing. Besides, it validates MRC software against various linguistic properties to give a specific and in-depth picture on linguistic capabilities of MRC software. Comprehensive experimental results show that our method can successfully reveal violations to the target linguistic properties without the labels. Moreover, it can reveal problems that have been concealed by the traditional validation. Comparison according to the properties provides deeper and more concrete ideas about different language understanding capabilities of the MRC software.

MET is a relatively new workshop on metamorphic testing for academic researchers and industry practitioners. The first international workshop on MET (MET 2016) was co-located with the 38th International Conference on Software Engineering (ICSE 2016) in Austin TX, USA on May 16, 2016. Since then the workshop has become an annual event. This paper reports on the fourth International Workshop on Metamorphic Testing (MET 2019) held in Montréal, Canada on May 26, 2019, as part of the 41st International Conference on Software Engineering (ICSE 2019). We first outline the aims of the workshop, followed by a discussion of its keynote speech and technical program.

Pyridine Nucleotide-Disulphide Oxidoreductase Domain 2 (PYROXD2), a Hepatitis B virus X protein (HBx)-interacting protein, is significantly down-regulated in hepatocellular carcinoma (HCC), however its exact biological function remains unclear. The aim of this study is to investigate the subcellular localization and biological function of PYROXD2 in hepatic cells. The results showed that PYROXD2 was imported to the mitochondrial inner membrane/matrix by Tom40 and Tim23, but not Mia40. PYROXD2 151-230aa might be the mitochondrial targeting sequence. PYROXD2 interacted with complex IV subunit COX5B. Knockout of PYROXD2 decreased MMP, intracellular ROS, complex IV activity, cell proliferation, ATP content and mtDNA copy number, but increased mtROS levels and the number of immature mitochondria. In summary, our data illustrated that PYROXD2 localizes to the mitochondrial inner membrane/matrix, and it plays important roles in regulating mitochondrial function.

GitHub is a popular code platform that provides infrastructures to facilitate collaborative development. A Pull Request (PR) is one of the key ideas to support collaboration. Developers are encouraged to submit PRs to ask for the integration of their contributions. In practice, not all submitted PRs can be integrated into the codebase by project maintainers. Existing studies have investigated factors affecting PR integration. Nevertheless, the code style of PRs, which is largely considered by project maintainers, has not been deeply studied yet. In this paper, we performed an exploratory analysis on the effect of code style on PR integration in GitHub. We modeled the code style via the inconsistency between a submitted PR and the existing code in its target codebase. Such modeling makes our study not limited by a specific definition of code style. We conducted our experiments on 50,092 closed PRs in 117 Java projects. Our findings show that: (1) There indeed exists code style inconsistency between PRs and the codebase. (2) Several code style criteria on how to use spaces or indents, make comments, and write code lines with a suitable length, tend to show more inconsistency among PRs. (3) A PR that is consistent with the current code style tends to be merged into the codebase more easily. (4) A PR that violates the current code style is likely to take more time to get closed. Our study shows evidence to developers about how to deliver better contributions to facilitate efficient collaboration.

Given a stack trace reported at the time of software crash, crash localization aims to pinpoint the root cause of the crash. Crash localization is known as a time-consuming and labor-intensive task. Without tool support, developers have to spend tedious manual effort examining a large amount of source code based on their experience. In this paper, we propose an automatic approach, namely CraTer, which predicts whether a crashing fault resides in stack traces or not (referred to as predicting crashing fault residence). We extract 89 features from stack traces and source code to train a predictive model based on known crashes. We then use the model to predict the residence of newly-submitted crashes. CraTer can reduce the search space for crashing faults and help prioritize crash localization efforts. Experimental results on crashes of seven real-world projects demonstrate that CraTer can achieve an average accuracy of over 92%.

Combining spectrum-based fault localization (SBFL) with other techniques is generally regarded as a feasible approach as advantages from both techniques would be preserved. Sendys which combines SBFL with slicing-hitting-set-computation is one of the promising techniques. However, all current evaluations on Sendys were obtained via empirical studies, which have inevitable threats to validity. Besides, purely empirical studies cannot reveal the essential reason that why Sendys performs well or badly, and whether all the complicated computations are necessary. Therefore, in this paper, we provide an in-depth theoretical analysis on Sendys , which can give definite and convincing conclusions. We generalize our previous theoretical framework on SBFL, to make it applicable to combined techniques like Sendys . We first provide a variant of current Sendys by patching its loophole of ignoring “zero or negative risk values” in normalization. This variant plays as a substitution of the original Sendys , as well as one of the baselines in our analysis. Then, by modifying a few steps of this variant, we propose an enhanced Sendys and theoretically prove its superiority over several other methods in single-fault scenario. Moreover, we provide a short-cut reformulation of the enhanced Sendys by preserving its performance, but only requiring very simple computations. And it is proved to be even better than traditional SBFL maximal formulas. As a complementary, our empirical studies with 13 subject programs demonstrate the obvious superiority of the enhanced Sendys , as well as its stability across different formulas in single-fault scenario. For multiple-fault cases, the variant of Sendys is observed to have the best performance. Besides, this variant has shown great helpfulness in improving the bad performance of the original Sendys when encountering the NOR problem.