Retention Prediction Models Research Articles

Retention time prediction for post-translationally modified peptides: Ser, Thr, Tyr-phosphorylation

The development of a peptide retention prediction model for reversed-phase chromatography applications in proteomics is reported for peptides carrying phosphorylated Ser, Thr and Tyr-residues. The major retention features have been assessed using a collection of over 10,000 phosphorylated/non-phosphorylated peptide pairs identified in a series 1D and 2D LC-MS/MS acquisitions using formic acid as ion pairing modifier. Single modification event on average results in increased peptide retention for phosphorylation of Ser (+ 1.46), Thr (+1.33), Tyr (+0.93% acetonitrile, ACN) on gradient elution scale for Luna C18(2) stationary phase. We established several composition and sequence specific features, which drive deviations from these average values. Thus, single phosphorylation of serine results in retention shifts ranging from -2.4 to 5.5% ACN depending on position of the residue, nature of nearest neighbour residues, peptide length, hydrophobicity and pI value, and its propensity to form amphipathic helical structures. We established that the altered ion-pairing environment upon phosphorylation is detrimental for this variability. Hydrophobicity of ion-pairing modifier directly informs the magnitude of expected shifts: (most hydrophilic) 0.5 % acetic acid (larger positive shift upon phosphorylation) > 0.1 % formic acid (positive) > 0.1 % trifluoroacetic (negative) > 0.1 % heptafluorobutyric acid (larger negative shift). The effect of phosphorylation has been also evaluated for several separation conditions used in the first dimension of 2D LC applications: high pH reversed-phase (RP), hydrophilic interaction liquid chromatography (HILIC), strong cation- and strong anion exchange separations.

Analysis of Octane Retention Prediction Model for Catalytic Cracked Gasoline Based on Ridge Regression Model and Gradient Descent Optimization

On the basis of the given material, in order to increase the RON retention of the catalytic cracking unit, the prediction model of gasoline octane retention and the best operation variable inversion model were established based on the Ridge regression model and Gradient descent method. First, based on the Ridge regression model, the leave-one method is used to obtain the relative importance of the operational variables, and select the most important variables, so as to reduce the characteristic dimension of the model; Then, the RON retention prediction model is trained based on the Ridge regression model; Finally, based on the trained Ridge regression model and its weight parameters, the optimal operating variables were optimized separately using the gradient when the operation variable has a range or no range of value. The experimental results show that when 146 are selected from 361 operating variables, the model loss value stabilizes; when α is 0.6, the test set R2 is 0.9882, test set MSE is 0.0193, and the comprehensive performance is better than the random forest, support vector machine model; When the operation variable has two categories of value range and no value range, 2,000 times, the best inversion value of the operation variable makes the RON retention prediction value of the test sample similar to the expected value, and the MAE drops to 2.89999�10-3 and 7.62939�10-6, respectively. In conclusion, the RON retention prediction model proposed in this study has good results, and the best operating variable can be reversed, based on the given material parameters, making the optimal RON retention quantity.

Compendium of Chromatographic Behavior of Post-translationally and Chemically Modified Peptides in Bottom-Up Proteomic Experiments.

We have systematically evaluated the chromatographic behavior of post-translationally/chemically modified peptides using data spanning over 70 of the most relevant modifications. These retention properties were measured for standard bottom-up proteomic settings (fully porous C18 separation media, 0.1% formic acid as ion-pairing modifier) using collections of modified/nonmodified peptide pairs. These pairs were generated by spontaneous degradation, chemical or enzymatic treatment, analysis of synthetic peptides, or the cotranslational incorporation of noncanonical proline analogues. In addition, these measurements were validated using external data acquired for synthetic peptides and enzymatically induced citrullination. Working in units of hydrophobicity index (HI, % ACN) and evaluating the average retention shifts (ΔHI) represent the simplest approach to describe the effect of modifications from a didactic point of view. Plotting HI values for modified (y-axis) vs nonmodified (x-axis) counterparts generates unique slope and intercept values for each modification defined by the chemistry of the modifying moiety: its hydrophobicity, size, pKa of ionizable groups, and position of the altered residue. These composition-dependent correlations can be used for coarse incorporation of PTMs into models for prediction of peptide retention. More accurate predictions would require the development of specific sequence-dependent algorithms to predict ΔHI values.

A multi-target QSRR approach to model retention times of small molecules in RPLC

Quantitative structure-retention relationship models (QSRR) have been utilized as an alternative to costly and time-consuming separation analyses and associated experiments for predicting retention time. However, achieving 100 % accuracy in retention prediction is unrealistic despite the existence of various tools and approaches. The limitations of vast data availability and time complexity hinder the use of most algorithms for retention prediction. Therefore, in this study, we examined and compared two approaches for modelling retention time using a dataset of small molecules with retention times obtained at multiple conditions, referred to as multi-targets (five pH levels: 2.7, 3.5, 5, 6.5, and 8 at gradient times of 20 min of mobile phase). The first approach involved developing separate models for predicting retention time at each condition (single-target approach), while the second approach aimed to learn a single model for predicting retention across all conditions simultaneously (multi-target approach). Our findings highlight the advantages of the multi-target approach over the single-target modelling approach. The multi-target models are more efficient in terms of size and learning speed compared to the single-target models. These retention prediction models offer two-fold benefits. Firstly, they enhance knowledge and understanding of retention times, identifying molecular descriptors that contribute to changes in retention behaviour under different pH conditions. Secondly, these approaches can be extended to address other multi-target property prediction problems, such as multi-quantitative structure Property(X) relationship studies (mt-QS(X)R).

Research on Customer Retention Prediction Model of VOD Platform Based on Machine Learning

Research on Customer Retention Prediction Model of VOD Platform Based on Machine Learning

Chromatographic behaviour of peptides modified with amine-reacting tags for relative protein quantitation in proteomic applications

Reversed-phase (RP) HPLC separation of peptides labeled with amine-reacting tags for relative protein quantitation (iTRAQ4, iTRAQ8 - isobaric tag for relative and absolute quantitation, TMT - tandem mass tag) has been investigated using large-scale proteomics derived retention datasets. These tags have similar chemistry but use linkers of different length and hydrophobicity, moving the positively charged functional groups further from peptide backbone. Peptide hydrophobicity (RP HPLC retention), on average, increases in the following order: non-labeled < iTRAQ4 < iTRAQ8 < TMT under both low pH (0.1% formic acid) and pH 10 eluent conditions. At the same time, the interplay between hydrophobicity and length of the labeling group drives the deviations from this order. Thus, longer and less hydrophobic iTRAQ8 moiety results in greater retention increase for peptides carrying amphipathic helical structures at the N-terminus. Development of a peptide retention prediction models for these modifications was achieved by predicting correspondent retention shifts ΔHI (hydrophobicity index,% acetonitrile) between unmodified and labelled peptide pairs.

Development of a retention prediction model in ion-pair reversed-phase HPLC for nucleoside triphosphates used as mRNA vaccine raw materials

The International Conference on Harmonization guidelines for quality on pharmaceutical development recommends a systematic development approach including robustness studies which assure performance of manufacturing and analytical method development of drug product. It was demonstrated that the retention prediction model for nucleoside triphosphates (NTPs) on ion-pair reversed-phase HPLC was developed by a highly accurate Kawabe’s model which supports the development of robust HPLC methods. As NTPs and its derivatives are typically used for Messenger ribonucleic acid (mRNA) vaccine production, adenosine-5′-triphosphate (ATP), guanosine-5′-triphosphate (GTP), cytidine-5′-triphosphate (CTP), 5-methylcytidine-5′-triphosphate (m5-CTP), uridine-5′-triphosphate (UTP), 5-methyluridine-5′-triphosphate (m5-UTP), pseudouridine-5′-triphosphate (Ψ-UTP) and N1-methylpseudouridine-5′-triphosphate (m1Ψ-UTP) were applied for prediction model development. By a comparison of the predicted retention factor in eight studied samples with the retention factor measured under six isocratic conditions, the absolute prediction error was 0.075 and also the prediction error (%) was 2.70%. In practical examples, analytical method for residual ATP, GTP, CTP, and m1Ψ-UTP in the commercial mRNA-based drugs and purity method for UTP derivatives were optimized by QbD approach. The design space for the minimum resolution between adjacent peaks was simulated with the models developed to evaluate the robustness of peak separation, and the optimal mobile phase condition was also simulated. As a conclusion, the desired peak was successfully separated under the optimized condition, and we thought that these retention models could optimize the mobile phase condition of the NTP analysis method for applying to various quality tests, such as quantity, purity and identity test for NTPs and its derivates in the mRNA-based drugs.

Peptide retention time prediction for peptides with post-translational modifications: N-terminal (α-amine) and lysine (ε-amine) acetylation

Development of a peptide retention prediction model in reversed-phase chromatography is reported for acetylated peptides – both N-terminal (α-) and side chain of Lys (ε-amine) residues. Large-scale proteomic 2D LC-MS analyses of acetylated/non-acetylated tryptic digest of whole human cell lysate have been used to assemble representative retention data sets of 25,000+ modified/non-modified pairs. This allowed elucidating chromatographic behaviour of modified peptides in three different separation modes: high pH reversed-phase, HILIC separation on amide phase (first dimension of 2D) and reversed-phase separation with formic acid as ion-pairing modifier in the second dimension. On average, N-terminal acetylation increases peptide RP retention at acidic pH by 5 Hydrophobicity Index units (% acetonitrile). Acetylation of first lysine adds another 4.1%. The magnitude of the retention shift varies greatly depending on the number of modified amines, peptide length, and N-terminal peptide sequence. Large retention shifts have been observed for peptides with hydrophobic N-termini and specifically peptides carrying sequences characteristic for amphipathic helical structures – all in complete agreement with major sequence-specific features of RP retention mechanism. The utility of the modified Sequence Specific Retention Calculator model has been verified for the in-vivo N-terminally acetylated peptides detected by 2D LC-MS/MS analysis of a yeast tryptic digest. The effect of N-terminal acetylation was also evaluated for six different HILIC columns, strong cation- and strong anion exchange separations using previously acquired 2D LC-MS/MS data.

Online English Teaching Course Score Analysis Based on Decision Tree Mining Algorithm

With the advent of the Big Data era, information and data are growing in spurts, fueling the deep application of information technology in all levels of society. It is especially important to use data mining technology to study the industry trends behind the data and to explore the information value contained in the massive data. As teaching and learning in higher education continue to advance, student academic and administrative data are growing at a rapid pace. In this paper, we make full use of student academic data and campus behavior data to analyze the data inherent patterns and correlations and use these patterns rationally to provide guidance for teaching activities and teaching management, thus further improving the quality of teaching management. The establishment of a data‐mining‐technology‐based college repetition warning system can help student management departments to strengthen supervision, provide timely warning information for college teaching management as well as leaders and counselors’ decision‐making, and thus provide early help to students with repetition warnings. In this paper, we use the global search advantage of genetic algorithm to build a GABP hybrid prediction model to solve the local minimum problem of BP neural network algorithm. The data validation results show that Recall reaches 95% and F1 result is about 86%, and the accuracy of the algorithm prediction results is improved significantly. It can provide a solid data support basis for college administrators to predict retention. Finally, the problems in the application of the retention prediction model are analyzed and corresponding suggestions are given.

Various aspects of retention index usage for GC-MS library search: A statistical investigation using a diverse data set

This work is devoted to the large-scale statistical evaluation of various aspects of using the retention index for GC-MS library search with a diverse data set. A search in a large library often does not give a correct compound even if a library contains it. One of the methods to improve a spectral library search procedure is to use the retention index information. The aim of this study is to explore some statistical peculiarities which can be helpful for development of automated software which uses a library search of diverse completely unknown compounds in a large database. A data set that was used in this work as a source of queries contains ~11 thousand spectra of compounds which belong to diverse chemical classes. Six equations for matching reference and experimental “retention index – spectrum” pairs were compared. It was found that good results can be obtained when a linear equation for similarity of pairs is used. Similarity of pairs is found as a sum of spectral similarity and of a product of a negative adjustable weight parameter and the absolute difference between reference and query retention indices. This equation performs equal or better than much more complex equations which contain two instead of one adjustable parameters. Widely used threshold-based approach, when candidates with high retention index deviation are rejected, performs worse than other equations. The use of predicted with neural networks retention indices as reference was also considered. Modern universal retention prediction models which are applicable to a wide variety of compounds are still quite inaccurate comparing with values from databases, but these predicted values allow to improve a library search as well. When predicted retention indices are used as reference, the linear equation for matching “retention index – spectrum” pairs also performs equal or better than other equations. The distribution of differences between query indices and reference indices (both calculated and experimental) was found close to exponential distribution near zero. The dependence of a fraction of correct identifications on the reference retention indices accuracy was studied. The addition of random noise with double exponential distribution to exact values was used to create “reference” retention indices with the predefined accuracy. The use of the molecular mass and molecular formula as additional constraints during a library search was also considered.

Application of Solvatic Model for Prediction of Retention at Gradient Elution in Reversed-Phase Liquid Chromatography

There are several different approaches for LC method development; beside traditional, different software programs for method development and optimization are available. The solvatic retention model of reversed-phase LC was applied for prediction of retention in the gradient elution mode for aripiprazole and its related substances described in European Pharmacopoeia. As some of these compounds have very similar and others quite different chemical structure, their separation is challenge. Prediction was suitable on examined stationary phases (C18, C8 and phenyl-hexyl) with 0.1% phosphoric acid as aqueous mobile phase and acetonitrile or methanol as organic modifier. Predicted retention times take into account structural formulae of compounds and properties of stationary and mobile phases result in average difference of 14-17% compared to experimental ones on phenyl-hexyl stationary phase, where the highest matching was obtained. After utilisation of the retention models with data from one experimental run, the average difference decrease to maximal 7% and after contribution of data from two experimental runs, to maximal 2%. For majority of studied compounds difference between predicted and experimental values on all examined stationary phases is lower than 3%.

Gradient boosting for the prediction of gas chromatographic retention indices

The estimation of gas chromatographic retention indices based on compounds structures is an importantproblem. Predicted retention indices can be used in a mass spectral library search for the identificationof unknowns. Various machine learning methods are used for this task, but methods based on decisiontrees, in particular gradient boosting, are not used widely. The aim of this work is to examine the usability ofthis method for the retention index prediction. 177 molecular descriptors computed with Chemistry Development Kit are used as the input representation of a molecule. Random subsets of the whole NIST 17 database are used as training, test and validation sets. 8000 trees with 6 leaves each are used. A neural network with one hidden layer (90 hidden nodes) is used for the comparison. The same data sets and the set of descriptors are used for the neural network and gradient boosting. The model based on gradient boosting outperforms the neural network with one hidden layer for subsets of NIST 17 and for the set of essential oils.The performance of this model is comparable or better than performance of other modern retention prediction models. The average relative deviation is ~3.0%, the median relative deviation is ~1.7% for subsets of NIST 17. The median absolute deviation is ~34 retention index units. Only non-polar liquid stationary phases (such as polydimethylsiloxane, 5% phenyl 95% polydimethylsiloxane, squalane) are considered. Errors obtained with different machine learning algorithms and with the same representation of the molecule strongly correlate with each other.

579. Machine-Learning Based Models for Prediction of Recurrence-free Catheter Retention After ALT Treatment of CLABSI in a Pediatric Population

BackgroundDeciding whether to attempt salvage of an infected central venous catheter (CVC) can be challenging. While line removal is the definitive treatment for central-line associated bloodstream infection (CLABSI), salvage may be attempted with systemic antibiotics and antibiotic lock therapy (ALT). Weighing risk and benefit of CVC salvage is limited by uncertainty in the future viability of salvaged CVCs. If a CVC is likely to require subsequent removal (e.g., due to recurrent infection) salvage may not be beneficial, whereas discarding a viable CVC is also not desirable. Here we describe a machine learning approach to predicting outcomes in CVC salvage.MethodsEpisodes of pediatric CLABSI cleared with ALT were identified by retrospective record review between January 1, 2008 and December 31, 2018 and were defined by a single positive central blood culture of a known pathogen or two matching cultures of a possible contaminant. Clearance was defined as 48-hours of negative cultures and relapse was defined as a matching positive blood culture after clearance. Predictive models [logistic regression (LR), random forest (RF), support vector machine (SVM) and an ensemble combining the three] were used to predict recurrence-free CVC retention (RFCR) at various time points using a training and test set approach.ResultsOverall, 712 instances CLABSI cleared with ALT were identified. Demographic and microbiological data are summarized in Tables 1 and 2. Few (8%) instances recurred in the first 28 days. 58% recurred at any time within the study period. Rates of RFCR were 75%, 43%, 22% and 10% at 28, 91, 182 and 365 days. Machine learning (ML) models varied in their ability to predict RFCR (Table 3). RF models performed best overall, although no model performed well at 91 days.ConclusionML models provide an opportunity to augment clinical decision making by learning patterns from data. In this case, estimating the likelihood of useful line retention in the future could help guide informed decisions on salvage vs. removal of infected CVCs. Limitations include the heterogeneity of clinical data and the use of an outcome capturing both clinical decision making (line removal) and infection recurrence. With further model development and prospective validation, practical machine learning models may prove useful to clinicians. Disclosures All authors: No reported disclosures.

Peptide retention time prediction for immobilized artificial membrane phosphatidylcholine stationary phase: method development and preliminary observations

Development of the first peptide retention prediction model for immobilized artificial membrane phosphatidylcholine (IAM.PC) stationary phase is reported. 2D LC-MS/MS analysis of a whole cell lysate of S. cerevisiae yielded a retention dataset of ~29,500 tryptic peptides; sufficient for confident assignment of retention coefficients which determine the contribution of individual amino acids in peptide retention. Retention data from the first dimension was used for the modeling: IAM.PC DD2 column, pH 7.4 ammonium bicarbonate, and water/acetonitrile gradient. Peptide separation using IAM.PC was compared to a standard C18 phase (Luna C18(2)). There was a significant reduction in peptide retention (~14% acetonitrile on average), indicating that the phosphatidylcholine stationary phase is significantly more hydrophilic. In comparison to the C18 phase, we found a substantial increase in the relative retention contribution for the positively charged Arg and Lys, and the aromatic Tyr, Trp and His residues. We also observed a decrease in retention contribution for the negatively charged Asp and Glu. This indicates an involvement of electrostatic interactions with the glycerophosphate functional groups, and possibly, delocalization effects from a hydrogen bond between the phosphate group and aromatic side chains in the separation mechanism.

Performance of combined fragmentation and retention prediction for the identification of organic micropollutants by LC-HRMS.

In nontarget screening, structure elucidation of small molecules from high resolution mass spectrometry (HRMS) data is challenging, particularly the selection of the most likely candidate structure among the many retrieved from compound databases. Several fragmentation and retention prediction methods have been developed to improve this candidate selection. In order to evaluate their performance, we compared two in silico fragmenters (MetFrag and CFM-ID) and two retention time prediction models (based on the chromatographic hydrophobicity index (CHI) and on log D). A set of 78 known organic micropollutants was analyzed by liquid chromatography coupled to a LTQ Orbitrap HRMS with electrospray ionization (ESI) in positive and negative mode using two fragmentation techniques with different collision energies. Both fragmenters (MetFrag and CFM-ID) performed well for most compounds, with average ranking the correct candidate structure within the top 25% and 22 to 37% for ESI+ and ESI- mode, respectively. The rank of the correct candidate structure slightly improved when MetFrag and CFM-ID were combined. For unknown compounds detected in both ESI+ and ESI-, generally positive mode mass spectra were better for further structure elucidation. Both retention prediction models performed reasonably well for more hydrophobic compounds but not for early eluting hydrophilic substances. The log D prediction showed a better accuracy than the CHI model. Although the two fragmentation prediction methods are more diagnostic and sensitive for candidate selection, the inclusion of retention prediction by calculating a consensus score with optimized weighting can improve the ranking of correct candidates as compared to the individual methods. Graphical abstract Consensus workflow for combining fragmentation and retention prediction in LC-HRMS-based micropollutant identification.

Sequence-Specific Model for Peptide Retention Time Prediction in Strong Cation Exchange Chromatography.

The development of a peptide retention prediction model for strong cation exchange (SCX) separation on a Polysulfoethyl A column is reported. Off-line 2D LC-MS/MS analysis (SCX-RPLC) of S.cerevisiae whole cell lysate was used to generate a retention dataset of ∼30 000 peptides, sufficient for identifying the major sequence-specific features of peptide retention mechanisms in SCX. In contrast to RPLC/hydrophilic interaction liquid chromatography (HILIC) separation modes, where retention is driven by hydrophobic/hydrophilic contributions of all individual residues, SCX interactions depend mainly on peptide charge (number of basic residues at acidic pH) and size. An additive model (incorporating the contributions of all 20 residues into the peptide retention) combined with a peptide length correction produces a 0.976 R2 value prediction accuracy, significantly higher than the additive models for either HILIC or RPLC. Position-dependent effects on peptide retention for different residues were driven by the spatial orientation of tryptic peptides upon interaction with the negatively charged surface functional groups. The positively charged N-termini serve as a primary point of interaction. For example, basic residues (Arg, His, Lys) increase peptide retention when located closer to the N-terminus. We also found that hydrophobic interactions, which could lead to a mixed-mode separation mechanism, are largely suppressed at 20-30% of acetonitrile in the eluent. The accuracy of the final Sequence-Specific Retention Calculator (SSRCalc) SCX model (∼0.99 R2 value) exceeds all previously reported predictors for peptide LC separations. This also provides a solid platform for method development in 2D LC-MS protocols in proteomics and peptide retention prediction filtering of false positive identifications.

Peptide Retention Time Prediction in Hydrophilic Interaction Liquid Chromatography: Data Collection Methods and Features of Additive and Sequence-Specific Models.

The development of a peptide retention prediction model for hydrophilic interaction liquid chromatography (XBridge Amide column) is described for a collection of ∼40 000 tryptic peptides. Off-line 2D LC-MS/MS analysis (HILIC-RPLC) of S. cerevisiae whole cell lysate has been used to acquire retention information for a HILIC separation. The large size of the optimization data set (more than 2 orders of magnitude compared to previous reports) permits the accurate assignment of hydrophilic retention coefficients of individual amino acids, establishing both the effects of amino acid position relative to peptide termini and the influence of peptide secondary structure in HILIC. The accuracy of a simple additive model with peptide length correction (R2 value of ∼0.96) was found to be much higher compared to an algorithm of similar complexity applied to RPLC (∼0.91). This indicates significantly smaller influence of peptide secondary structure and interactions with counterions in HILIC. Nevertheless, sequence-specific features were found. Helical peptides that tend to retain stronger than predicted in RPLC exhibit negative prediction errors using an additive HILIC model. N-cap helix stabilizing motifs, which increase retention of amphipathic helical peptides in RP, reduce peptide retention in HILIC independently of peptide amphipathicity. Peptides carrying multiple Pro and Gly (residues with lowest helical propensity) retain stronger than predicted. We conclude that involvement of the peptide backbone's carbonyl and amide groups in hydrogen-bond stabilization of helical structures is a major factor, which determines sequence-dependent behavior of peptides in HILIC. The incorporation of observed sequence dependent features into our Sequence-Specific Retention Calculator HILIC model resulted in 0.98 R2 value correlation, significantly exceeding the predictive performance of all RP and HILIC models developed for complex mixtures of tryptic peptides.

The Separation and Quantitation of Peptides with and without Oxidation of Methionine and Deamidation of Asparagine Using Hydrophilic Interaction Liquid Chromatography with Mass Spectrometry (HILIC-MS).

Peptides with deamidated asparagine residues and oxidized methionine residues are often not resolved sufficiently to allow quantitation of their native and modified forms using reversed phase (RP) chromatography. The accurate quantitation of these modifications is vital in protein biotherapeutic analysis because they can affect a protein's function, activity, and stability. We demonstrate here that hydrophilic interaction liquid chromatography (HILIC) adequately and predictably separates peptides with these modifications from their native counterparts. Furthermore, coefficients describing the extent of the hydrophilicity of these modifications have been derived and were incorporated into a previously made peptide retention prediction model that is capable of predicting the retention times of peptides with and without these modifications. Graphical Abstract ᅟ.

Comparison of peptide retention prediction algorithm in reversed-phase chromatography. Comment on "Predictive chromatography of peptides and proteins as a complementary tool for proteomics", by I. A. Tarasova, C. D. Masselon, A. V. Gorshkov and M. V. Gorshkov, Analyst, 2016, 141, 4816.

In a recent mini-review, Tarasova et al. (Analyst, 2016, 141, 4816) compared several peptide retention prediction models for reversed-phase HPLC developed for applied proteomics. The mechanism of peptide retention in RP-HPLC is very complex. Peptide separation selectivity is affected by multiple parameters of a separation system. Consequently, proper comparison of the models requires a detailed knowledge (to provide the best match) of separation conditions used to acquire both optimization and test datasets. Sequence-specific retention calculator - a benchmark tool in this field - consistently provides 0.96-0.965 R2-value correlations between experimental and predicted retention values, when applied correctly to extremely complex mixtures of peptides.

Implementing radial basis function neural network for prediction of surfactant retention in petroleum production and processing industries

ABSTRACTChemical flooding is an effective way to gain higher oil recovery as part of a tertiary oil recovery scheme. There are several variables contribute in surfactant retention in petroleum production including type of rock, pH, chemical structure of surfactant, salinity of formation water, acidity of oil, mobility, microemulsion viscosity, and cosolvent concentration. Although different theoretical studies on the mechanisms of surfactant retention are reported in the literature there is little research on the development of an accurate and effective model for prediction of surfactant retention in petroleum production. In this study, radial basis function was developed based on experimental dynamic surfactant retention data. The experimental data include a wide range of conditions. Results of the modeling study showed that the developed model is very accurate and robust in prediction of actual surfactant retention data. In addition, the comparison between the proposed model in this study and available models in literature showed the superiority of this model.