Robotic search for optimal cell culture in regenerative medicine.

Induced differentiation is one of the most experience- and skill-dependent experimental processes in regenerative medicine, and establishing optimal conditions often takes years. We developed a robotic AI system with a batch Bayesian optimization algorithm that autonomously induces the differentiation of induced pluripotent stem cell-derived retinal pigment epithelial (iPSC-RPE) cells. From 200 million possible parameter combinations, the system performed cell culture in 143 different conditions in 111 days, resulting in 88% better iPSC-RPE production than that obtained by the pre-optimized culture in terms of the pigmentation scores. Our work demonstrates that the use of autonomous robotic AI systems drastically accelerates systematic and unbiased exploration of experimental search space, suggesting immense use in medicine and research.

A Variable Scheduling Maintenance Culture Platform for Mammalian Cells.

Cell culturing is a basic experimental technique in cell biology and medical science. However, culturing high-quality cells with a high degree of reproducibility relies heavily on expert skills and tacit knowledge, and it is not straightforward to scale the production process due to the education bottleneck. Although many automated culture systems have been developed and a few have succeeded in mass production environments, very few robots are permissive of frequent protocol changes, which are often required in basic research environments. LabDroid is a general-purpose humanoid robot with two arms that performs experiments using the same tools as humans. Combining our newly developed AI software with LabDroid, we developed a variable scheduling system that continuously produces subcultures of cell lines without human intervention. The system periodically observes the cells on plates with a microscope, predicts the cell growth curve by processing cell images, and decides the best times for passage. We have succeeded in developing a system that maintains the cultures of two HEK293A cell plates with no human intervention for 192 h.

Identification of endocrine disruptor biodegradation by integration of structure-activity relationship with pathway analysis.

We present a SAR method that can predict estrogen-like endocrine disrupting chemical (EDC) activity as well as key biodegradation steps for detoxification. This method is based on a recent graph-mining algorithm developed by Kudo et al., which generates a set of descriptors from all potent chemical fragments (including rings). This method is novel in that it achieves chemical diversity in the training data set by sampling another data set of larger diversity. The model achieved an 83% accuracy prediction rate, and identified 1291 EDC candidates from the KEGG database. From this set of candidate compounds, bisphenol A was chosen for assay validation and biodegradation pathway analysis. Results showed that bisphenol A exhibited estrogen-like activity and was degraded in three distinct reactions. The prediction model provided information on the mechanism of the ligand-target binding, such as key functional groups involved. We focused on the enzyme commission number, which is useful for analyses of biodegradation pathways. Results identified oxygenases, ether hydrolases, and carbon-halide lyases as being important in the biodegradation pathway. This combined approach provided new information regarding the biodegradation of EDCs, and can potentially be extended to applications with transcriptomic, proteomic, and metabolomic data to provide a quick screen of biological activity and biodegradation pathway(s).

Mining frequent stem patterns from unaligned RNA sequences.

MOTIVATION: In detection of non-coding RNAs, it is often necessary to identify the secondary structure motifs from a set of putative RNA sequences. Most of the existing algorithms aim to provide the best motif or few good motifs, but biologists often need to inspect all the possible motifs thoroughly. RESULTS: Our method RNAmine employs a graph theoretic representation of RNA sequences and detects all the possible motifs exhaustively using a graph mining algorithm. The motif detection problem boils down to finding frequently appearing patterns in a set of directed and labeled graphs. In the tasks of common secondary structure prediction and local motif detection from long sequences, our method performed favorably both in accuracy and in efficiency with the state-of-the-art methods such as CMFinder. AVAILABILITY: The software is available upon request.

Use of morphological analysis in protein name recognition.

Protein name recognition aims to detect each and every protein names appearing in a PubMed abstract. The task is not simple, as the graphic word boundary (space separator) assumed in conventional preprocessing does not necessarily coincide with the protein name boundary. Such boundary disagreement caused by tokenization ambiguity has usually been ignored in conventional preprocessing of general English. In this paper, we argue that boundary disagreement poses serious limitations in biomedical English text processing, not to mention protein name recognition. Our key idea for dealing with the boundary disagreement is to apply techniques used in Japanese morphological analysis where there are no word boundaries. Having evaluated the proposed method with GENIA corpus 3.02, we obtain F-measure of 69.01 on a strict criterion and 79.32 on a relaxed criterion. The result is comparable to other published work in protein name recognition, without resorting to manually prepared ad hoc feature engineering. Further, compared to the conventional preprocessing, the use of morphological analysis as preprocessing improves the performance of protein name recognition and reduces the execution time.

A new serum antibody test kit (E plate) for evaluation of Helicobacter pylori eradication.

OBJECTIVE: Serological antibody test have been widely performed to detect the presence of H. pylori, but they have not been used to evaluate the status of H. pylori after eradication. In this study we evaluated the diagnostic accuracy of a new serological test kit (E-plate) after eradication. METHOD: Eradication of H. pylori was performed in 100 patients by proton pump inhibitor (PPI)+amoxicillin (AMPC)+clarithromycin (CAM) or PPI+AMPC therapy. Evaluation of H. pylori was done by culture, histology and rapid urease test before, and 8 weeks after, the treatment. Serological tests were also performed before and after treatment using the E plate. Cure was defined as no evidence of H. pylori at 8 weeks after the treatment. Receiver operating characteristic (ROC) analysis was performed to determine the ideal cut-off value for percentage change in the serological test. RESULT: Success was obtained in 73 patients, failure in 20 patients and there were 7 dropouts. Serological test value was significantly decreased after treatment (44.3 +/- 29.6 U/ ml) compared to before treatment (94.8 +/- 73.2 U/ml) in the successful cases. In contrast, those with no significant change after treatment (62.7 +/- 31.3 U/ml) compared to before treatment (72.9 +/- 47.7 U/ml) were considered as failure cases. ROC analysis revealed that cut-off values of a 20%, 30%, and 40% decrease on E plate result yielded a sensitivity of 95.5%, 92.4%, 71.2% and a specificity of 73.3%, 84.2%, 94.7%, respectively. CONCLUSION: The new E plate serological test kit for H. pylori was useful for distinguishing success from failure 8 weeks after completion of eradication therapy for H. pylori.