Development of an experiment-split method for benchmarking the generalization of a PTM site predictor: Lysine methylome as an example

Many computational classifiers have been developed to predict different types of post-translational modification sites. Their performances are measured using cross-validation or independent test, in which experimental data from different sources are mixed and randomly split into training and test sets. However, the self-reported performances of most classifiers based on this measure are generally higher than their performances in the application of new experimental data. It suggests that the cross-validation method overestimates the generalization ability of a classifier. Here, we proposed a generalization estimate method, dubbed experiment-split test, where the experimental sources for the training set are different from those for the test set that simulate the data derived from a new experiment. We took the prediction of lysine methylome (Kme) as an example and developed a deep learning-based Kme site predictor (called DeepKme) with outstanding performance. We assessed the experiment-split test by comparing it with the cross-validation method. We found that the performance measured using the experiment-split test is lower than that measured in terms of cross-validation. As the test data of the experiment-split method were derived from an independent experimental source, this method could reflect the generalization of the predictor. Therefore, we believe that the experiment-split method can be applied to benchmark the practical performance of a given PTM model. DeepKme is free accessible via https://github.com/guoyangzou/DeepKme.

Predicting financial distress in the Indian textile sector

The purpose of this paper is to predict the financial distress of companies of the Indian textile sector using Altman Z score. The analysis conducted on 161 listed textile companies in India for a period of 10 years from 2009 to 2018. All the listed companies are categorized into large, medium, and small using the median split method based on the size of total assets. Kruskal Wallis test is applied to test whether the mean z-score is different for each category of companies. This research study shows that majority of the companies in the Indian textile sector are facing financial distress. Further, it shows that the z score of small, medium, and large-scale textile companies in India is significantly different.

Slower adaptation of control strategies in individuals with high impulsive tendencies

Flexible use of reactive and proactive control according to environmental demands is the key to adaptive behavior. In this study, forty-eight adults performed ten blocks of an AX-CPT task to reveal the strength of proactive control by the calculation of the proactive behavioral index (PBI). They also filled out the UPPS questionnaire to assess their impulsiveness. The median-split method based on the global UPPS score distribution was used to categorize participants as having high (HI) or low (LI) impulsiveness traits. The analyses revealed that the PBI was negatively correlated with the UPPS scores, suggesting that the higher is the impulsiveness, the weaker the dominance of proactive control processes. We showed, at an individual level, that the PBI increased across blocks and suggested that this effect was due to a smaller decrease in reactive control processes. Notably, the PBI increase was slower in the HI group than in the LI group. Moreover, participants who did not adapt to task demands were all characterized as high impulsive. Overall, the current study demonstrates that (1) impulsiveness is associated with less dominant proactive control due to (2) slower adaptation to task demands (3) driven by a stronger reliance on reactive processes. These findings are discussed in regards to pathological populations.

AFIKS DERIVASIONAL BER- PADA MEDIA MASSA INDONESIA

News is part of the mass media that is easily accessible to all people. News has the characteristics of language is short, concise and objective. The use of language in writing news must pay attention to the writing of word structures, word formation in accordance with PUEBI. One of the word formation that occurs is the result of affixation process. One of the productive affixes in Indonesian is the prefix ber-. This study discusses the derivational affixes contained in mass media coverage in Indonesia. Data sources came from national, regional, and local online mass media during the period of September 2020. Data were obtained using observation methods and note-taking techniques. Data were analyzed using the split method with direct element sharing techniques. The use of the affix on the Indonesian mass media in the form of Ber+N= V has the meaning (1) in the state; (2) obtain, produce; (3) have; and (4) reflexive. Whereas Ber+A= V means 'in a state'. Ber+Adv.= V means 'in a state'. Ber+N= Num. has the meaning of 'collection / collective', and Ber+bound morpheme= V has the meaning (1) 'in a state'; (2) 'have'. Variations in the use of affixes are mostly carried out in the national media and the affixes of ber- which are not yet available in the Kridalaksana book are Ber+Adv.= V, Ber+N= Num. , and Ber+bound morpheme= V.

The Comparison of the Formation of Indonesian and Arabic Plural Meanings (Contrastive Analysis)

This study compares the process of forming plural words in Indonesian and Arabic to determine the similarities and differences in the markers of the two languages’ plural meaning through contrastive analysis studies. The data in this study are fragments of speech that are thought to contain plural meaning. They are taken from the source of data in the form of speech that contains plural meanings in the two languages’ books. The data are analyzed using the split method with the Direct Element Dividing technique and the equivalent method using the Determination Element Determination technique. This study indicates that the formation of plural languages in Indonesian and Arabic has similarities and differences in rules. The equation includes the addition of affixes, numeral, plural lexical meanings, and changes in pronouns. The differences in the rules of the formation of plural words of Indonesian and Arabic include reduplication, internal modification, the existence of dualistic forms, gender differences in words, case differences, and subject-predicate relationships.

The Evaluation of the Corrosion Rates of Alloys Applied to the Heating Tower Heat Pump (HTHP) by Machine Learning

The corrosion rate is an important indicator describing the degree of metal corrosion, and quantitative analysis of the corrosion rate is of great significance. In the present work, the support vector machine (SVM) and the artificial neural network (ANN) integrating the k-fold split method and the root-mean-square prop (RMSProp) optimizer are used to evaluate the corrosion rates of alloys, i.e., copper H65, aluminum 3003, and 20# steel, applied to the heating tower heat pump (HTHP) in various anti-freezing solutions at different corrosion times, flow velocities, and temperatures. The mean-square error (MSE) versus the epoch of the ANN model shows that the result breaks the local minimum and is at or close to the global minimum. Comparisons of the SVM-/ANN-evaluated corrosion rates and the measured ones show good agreements, demonstrating the good reliability of the obtained SVM and ANN models. Moreover, the ANN model is recommended since it performs better than the SVM model according to the obtained R2 value. The present work can be further applied to predicting the corrosion rate without any prior experiment for improving the service life of the HTHP.

Exploring an experiment-split method to estimate the generalization ability in new data: DeepKme as an example

A Large number of predictors have been built based on different data sets for predicting different post-translational modification sites. However, limited to our knowledge, most of them gave an overfitting estimation of their generalization ability in new data because of the intrinsic trait—not considering the experimental sources of the new data—of the cross-validation method. Thus, we proposed and explored a new method—the experiment-split method—imitating the blinded assessment to deal with the overfitting problem in the new data. The experiment-split method logically split the training and test data based on the data’s different experimental sources, and the new data can be regarded as the data from different experimental sources. To specifically illustrate the experiment-split method, we combined an actual application, DeepKme—a predictor built by us for the lysine methylation sites, to demonstrate how it be used in the true scenarios. We compared the cross-validation method with the experiment-split method. The result suggested the experiment-split method could effectively relieve the overfitting compared with the cross-validation method and may be widely used in the field of identification participated by multiple experiments. We believe DeepKme would facilitate the related researchers’ deep thought of the experiment-split method and the overfitting phenomenon, and of course, advance the study of the lysine methylation and similar fields.