Objective:This study aims to analyze the performance of five automatic cell type annotation methods in single cell RNA sequencing (scRNA-seq) data.Methods:Simulated data were generated using the Splatter package in R language, taking into account two data characteristics: the number of cells and the number of genes. The actual data came from the GSE10245 scRNA seq dataset of non-small cell lung cancer in Gene Expression Omnibus (GEO) database, the data had been pre-processed and batch effects had been eliminated. The automatic cell type recognition (ACTINN) of neural networks, the single-cell type annotation method based on deep learning (scDeepSort), the reference batch transcriptome annotation scRNA seq R-package (SingleR), the cross platform and cross species scRNA seq data classifier (SingleCellNet), and the cross scRNA seq dataset projection (scMap-cell) were implemented using the Tensorflow library in Python. The performance evaluation indicators for cell type annotation included accuracy (ACC), F1-score, and Matthews correlation coefficient (MCC). Each method was validated using ten fold cross validation, and the average value was taken after 50 repeated runs for performance comparison between methods. The Dunnett's t-test in the DescTools package of R language was used for multiple comparisons between ACTINN and other four methods. Results:Under 12 different scenarios (3 levels of cell numbers × 4 levels of gene numbers), simulated data analysis showed that compared with scDeepSort, SingleR, SingleCellNet, and scMap-cell, the percentage increase in ACC value of ACTINN ranged from 3.31% to 14.59%, 1.38% to 13.03%, 12.98% to 25.25%, and 20.72% to 29.62%, respectively; the range of F1 score improvement percentages were 2.75% - 22.74%, 2.46% - 23.68%, 5.07% - 27.47%, and 10.27% - 31.47%, respectively; the percentage increase ranges for MCC values were 3.42% - 9.75%, 2.26% - 7.61%, 5.41% - 11.11%, and 8.27% - 15.22%, respectively. Actual data analysis showed that the ACC value of ACTINN was 81.0%, which was increased by 2.1%, 5.2%, 7.9%, and 8.9% compared with the above four methods, respectively; the F1-score value was 80.5%, which was increased by 2.3%, 5.9%, 2.4%, and 6.0%, respectively; the MCC value was 83.3%, which was increased by 0.9%, 2.5%, 3.4%, and 11.2%, respectively. The results of Dunnett's t-test showed that the difference was not statistically significant in ACC values between scDeepSort and ACTINN ( P = 0.821), in F1-score values between scDeepSort and ACTINN ( P = 0.498), and in MCC values between scDeepSort, SingleCellNet and ACTINN ( P = 0.904, 0.134). However, the differences were statistically significant in other multiple comparisons ( P < 0.05). Conclusions:ACTINN and scDeepSort have good performance in cell type annotation, with ACTINN showing outstanding performance and SingleR showing robust performance, while SingleCellNet and scMap-cell have relatively limited performance. This suggests that self-attention mechanism algorithm based on Transformer framework is expected to promote further development of automatic cell annotation methods.

Objective:This study aims to analyze the performance of five automatic cell type annotation methods in single cell RNA sequencing (scRNA-seq) data.Methods:Simulated data were generated using the Splatter package in R language, taking into account two data characteristics: the number of cells and the number of genes. The actual data came from the GSE10245 scRNA seq dataset of non-small cell lung cancer in Gene Expression Omnibus (GEO) database, the data had been pre-processed and batch effects had been eliminated. The automatic cell type recognition (ACTINN) of neural networks, the single-cell type annotation method based on deep learning (scDeepSort), the reference batch transcriptome annotation scRNA seq R-package (SingleR), the cross platform and cross species scRNA seq data classifier (SingleCellNet), and the cross scRNA seq dataset projection (scMap-cell) were implemented using the Tensorflow library in Python. The performance evaluation indicators for cell type annotation included accuracy (ACC), F1-score, and Matthews correlation coefficient (MCC). Each method was validated using ten fold cross validation, and the average value was taken after 50 repeated runs for performance comparison between methods. The Dunnett's t-test in the DescTools package of R language was used for multiple comparisons between ACTINN and other four methods. Results:Under 12 different scenarios (3 levels of cell numbers × 4 levels of gene numbers), simulated data analysis showed that compared with scDeepSort, SingleR, SingleCellNet, and scMap-cell, the percentage increase in ACC value of ACTINN ranged from 3.31% to 14.59%, 1.38% to 13.03%, 12.98% to 25.25%, and 20.72% to 29.62%, respectively; the range of F1 score improvement percentages were 2.75% - 22.74%, 2.46% - 23.68%, 5.07% - 27.47%, and 10.27% - 31.47%, respectively; the percentage increase ranges for MCC values were 3.42% - 9.75%, 2.26% - 7.61%, 5.41% - 11.11%, and 8.27% - 15.22%, respectively. Actual data analysis showed that the ACC value of ACTINN was 81.0%, which was increased by 2.1%, 5.2%, 7.9%, and 8.9% compared with the above four methods, respectively; the F1-score value was 80.5%, which was increased by 2.3%, 5.9%, 2.4%, and 6.0%, respectively; the MCC value was 83.3%, which was increased by 0.9%, 2.5%, 3.4%, and 11.2%, respectively. The results of Dunnett's t-test showed that the difference was not statistically significant in ACC values between scDeepSort and ACTINN ( P = 0.821), in F1-score values between scDeepSort and ACTINN ( P = 0.498), and in MCC values between scDeepSort, SingleCellNet and ACTINN ( P = 0.904, 0.134). However, the differences were statistically significant in other multiple comparisons ( P < 0.05). Conclusions:ACTINN and scDeepSort have good performance in cell type annotation, with ACTINN showing outstanding performance and SingleR showing robust performance, while SingleCellNet and scMap-cell have relatively limited performance. This suggests that self-attention mechanism algorithm based on Transformer framework is expected to promote further development of automatic cell annotation methods.

ObjectiveTo observe the effect of dressing and moxibustion-pretreatment on the local joint swelling and stress hormone in hypothalami of rats with adjuvant arthritis (AA) at early and secondary stages.MethodsForty Wistar rats were randomly divided into 5 groups: normal group, early and secondary model groups, early and secondary pre-dressing and moxibustion (PDM) groups. The dressing with Chinese herb and moxibustion was stuck on Dazhui point (GV14) before the AA model established. The effects of dressing and moxibustion-pretreatment on the feet swelling and corticotropin-releasing-hormone (CRH), beta-endorphin (β-EP) and neuropeptide-Y (NPY) in hypothalami were observed.ResultsThe right feet swelling rate at early and secondary stages obviously increased after modeling ( P<0.01), and it became lower in early and secondary PDM groups than in model groups at the same phases ( P<0.05～0.01). The level of hypothalamic CRH was higher after modeling ( P<0.05～0.01), compared with early model group, it had a tendency to going down in early PDM group, moreover, in secondary PDM group the level was similar with the normal group. The level of hypothalamic β-EP increased significantly in early model group ( P<0.01), and lightly changed in secondary model group, it decreased in early PDM group but increased in secondary PDM group, compared with model groups at the same stages ( P<0.05). The level of hypothalamic NPY increased significantly after modeling, and it declined in the secondary PDM group ( P<0.05).ConclusionDressing and moxibustion-pretreatment can relief feet swelling of AA rats, which may be related with its regulative effect on the level of hypothalamic CRH, β-EP and NPY.

ObjectiveTo explore the effect of dressing-pretreatment on forepart inflammation in adjuvant arthritis (AA) rats.MethodsAA model was established by subcutaneous injection of Freund's complete adjuvant into the right hind-paw (0.1 ml/animal). 24 Wistar rats were randomly divided into the control group, model group and dressing-pretreatment groups. In dressing-pretreatment group, "Dazhui" (GV14) was applied by a kind of dressing invented by the researchers. The foot swelling rate, serum concentrations of interleukin-1 (IL-1), IL-6 and tumor necrotic factor-α (TNF-α) were assessed.ResultsCompared with the control group, the foot swelling rate and serum concentrations of IL-1, IL-6 and TNF-α of the model group and dressing-pretreatment group increased significantly (P<0.01); but the foot swelling rate and serum concentrations of IL-1 and IL-6 in dressing-pretreatment group decreased significantly compared with those of the model group (P<0.05～0.01), and there was a downtrend in TNF-α. HE stain showed soakage of phlogistic cell in cartilage and arthritis membrane, hyperplasia of vein and breakage of cartilage were better in dressing-pretreatment group compared with that in the model group.ConclusionDressing-pretreatment can decrease forepart inflammation and pathological damage of arthritis in AA rats, and can prevent the harm of consequent disease to some extent.