The Experts below are selected from a list of 7197 Experts worldwide ranked by ideXlab platform
Marc D. Polanka - One of the best experts on this subject based on the ideXlab platform.
-
An Experimental Study of Combustor Exit Profile Shapes on Endwall Heat Transfer in High Pressure Turbine Vanes
Journal of Turbomachinery, 2009Co-Authors: M. D. Barringer, Karen A. Thole, Marc D. PolankaAbstract:The design and development of current and future gas turbine engines for aircraft propulsion have focused on operating the high pressure turbine at increasingly elevated temperatures and pressures. The drive toward thermal operating conditions near theoretical stoichiometric limits as well as increasingly stringent requirements on reducing harmful emissions both equate to the temperature Profiles exiting combustors and entering turbines becoming less peaked than in the past. This drive has placed emphasis on determining how different types of inlet temperature and pressure Profiles affect the first stage airfoil endwalls. The goal of the current study was to investigate how different radial Profiles of temperature and pressure affect the heat transfer along the vane endwall in a high pressure turbine. Testing was performed in the Turbine Research Facility located at the Air Force Research Laboratory using an inlet Profile Generator. Results indicate that the convection heat transfer coefficients are influenced by both the inlet pressure Profile shape and the location along the endwall. The heat transfer driving temperature for inlet Profiles that are nonuniform in temperature is also discussed.
-
Migration of Combustor Exit Profiles Through High Pressure Turbine Vanes
Journal of Turbomachinery, 2009Co-Authors: M. D. Barringer, Karen A. Thole, Marc D. Polanka, John P. Clark, P. J. KochAbstract:The high pressure turbine stage within gas turbine engines is exposed to combustor exit flows that are nonuniform in both stagnation pressure and temperature. These highly turbulent flows typically enter the first stage vanes with significant spatial gradients near the inner and outer diameter endwalls. These gradients can result in secondary flow development within the vane passage that is different than what classical secondary flow models predict. The heat transfer between the working fluid and the turbine vane surface and endwalls is directly related to the secondary flows. The goal of the current study was to examine the migration of different inlet radial temperature and pressure Profiles through the high turbine vane of a modern turbine engine. The tests were performed using an inlet Profile Generator located in the Turbine Research Facility at the Air Force Research Laboratory. Comparisons of area-averaged radial exit Profiles are reported as well as Profiles at three vane pitch locations to document the circumferential variation in the Profiles. The results show that the shape of the total pressure Profile near the endwalls at the inlet of the vane can alter the redistribution of stagnation enthalpy through the airfoil passage significantly. Total pressure loss and exit flow angle variations are also examined for the different inlet Profiles.
-
Effects of Combustor Exit Profiles on Vane Aerodynamic Loading and Heat Transfer in a High Pressure Turbine
Journal of Turbomachinery, 2009Co-Authors: M. D. Barringer, Karen A. Thole, Marc D. PolankaAbstract:The flow and thermal fields exiting gas turbine combustors dictate the overall performance of the downstream turbine. The goal of this work was to investigate the effects of engine representative combustor exit Profiles on high pressure turbine vane aerodynamics and heat transfer. The various Profiles were produced using a nonreacting turbine inlet Profile Generator in the Turbine Research Facility (TRF) located at the Air Force Research Laboratory (AFRL). This paper reports how the pressure loading and heat transfer along the vane surface was affected by different turbine inlet pressure and temperature Profiles at different span locations. The results indicate that the inlet total pressure Profiles affected the aerodynamic loading by as much as 10%. The results also reveal that the combination of different total pressure and total temperature Profiles significantly affected the vane heat transfer relative to a baseline test with uniform inlet total pressure and total temperature. Near the inner diameter endwall, the baseline heat transfer was reduced 30‐40% over the majority of the vane surface. Near the outer dimeter endwall, it was found that certain inlet Profiles could increase the baseline heat transfer by 10‐ 20%, while other Profiles resulted in a decrease in the baseline heat transfer by 25‐35%. This study also shows the importance of knowing an accurate prediction of the local flow driving temperature when determining vane surface heat transfer. DOI: 10.1115/1.2950051
-
Migration of Combustor Exit Profiles Through High Pressure Turbine Vanes
Volume 4: Turbo Expo 2007 Parts A and B, 2007Co-Authors: M. D. Barringer, Marc D. Polanka, John P. Clark, P. J. Koch, Karen A. TholeAbstract:The high pressure turbine stage within gas turbine engines is exposed to combustor exit flows that are nonuniform in both stagnation pressure and temperature. These highly turbulent flows typically enter the first stage vanes with significant spatial gradients near the inner and outer diameter endwalls. These gradients can result in secondary flow development within the vane passage that is different than what classical secondary flow models predict. The heat transfer between the working fluid and the turbine vane surface and endwalls is directly related to the secondary flows. The goal of the current study was to examine the migration of different inlet radial temperature and pressure Profiles through the high turbine vane of a modern turbine engine. The tests were performed using an inlet Profile Generator located in the Turbine Research Facility (TRF) at the Air Force Research Laboratory (AFRL). Comparisons of area-averaged radial exit Profiles are reported as well as Profiles at three vane pitch locations to document the circumferential variation in the Profiles. The results show that the shape of the total pressure Profile near the endwalls at the inlet of the vane can alter the redistribution of stagnation enthalpy through the airfoil passage significantly. Total pressure loss and exit flow angle variations are also examined for the different inlet Profiles.Copyright © 2007 by ASME
-
Experimental Evaluation of an Inlet Profile Generator for High-Pressure Turbine Tests
Journal of Turbomachinery, 2006Co-Authors: M. D. Barringer, Karen A. Thole, Marc D. PolankaAbstract:Improving the performance and durability of gas turbine aircraft engines depends highly on achieving a better understanding of the flow interactions between the combustor and turbine sections. The flow exiting the combustor is very complex and it is characterized primarily by elevated turbulence and large variations in temperature and pressure. The heat transfer and aerodynamic losses that occur in the turbine passages are driven primarily by these spatial variations. To better understand these effects, the goal of this work is to benchmark an adjustable turbine inlet Profile Generator for the Turbine Research Facility (TRF) at the Air Force Research Laboratory. The research objective was to experimentally evaluate the performance of the nonreacting simulator that was designed to provide representative combustor exit Profiles to the inlet of the TRF turbine test section. This paper discusses the verification testing that was completed to benchmark the performance of the Generator. Results are presented in the form of temperature and pressure Profiles as well as turbulence intensity and length scale. This study shows how a single combustor geometry can produce significantly different flow and thermal field conditions entering the turbine. Engine designers should place emphasis on obtaining accurate knowledge of the flow distribution within the combustion chamber. Turbine inlet conditions with significantly different Profile shapes can result in altered flow physics that can change local aerodynamics and heat transfer.
M. D. Barringer - One of the best experts on this subject based on the ideXlab platform.
-
An Experimental Study of Combustor Exit Profile Shapes on Endwall Heat Transfer in High Pressure Turbine Vanes
Journal of Turbomachinery, 2009Co-Authors: M. D. Barringer, Karen A. Thole, Marc D. PolankaAbstract:The design and development of current and future gas turbine engines for aircraft propulsion have focused on operating the high pressure turbine at increasingly elevated temperatures and pressures. The drive toward thermal operating conditions near theoretical stoichiometric limits as well as increasingly stringent requirements on reducing harmful emissions both equate to the temperature Profiles exiting combustors and entering turbines becoming less peaked than in the past. This drive has placed emphasis on determining how different types of inlet temperature and pressure Profiles affect the first stage airfoil endwalls. The goal of the current study was to investigate how different radial Profiles of temperature and pressure affect the heat transfer along the vane endwall in a high pressure turbine. Testing was performed in the Turbine Research Facility located at the Air Force Research Laboratory using an inlet Profile Generator. Results indicate that the convection heat transfer coefficients are influenced by both the inlet pressure Profile shape and the location along the endwall. The heat transfer driving temperature for inlet Profiles that are nonuniform in temperature is also discussed.
-
Migration of Combustor Exit Profiles Through High Pressure Turbine Vanes
Journal of Turbomachinery, 2009Co-Authors: M. D. Barringer, Karen A. Thole, Marc D. Polanka, John P. Clark, P. J. KochAbstract:The high pressure turbine stage within gas turbine engines is exposed to combustor exit flows that are nonuniform in both stagnation pressure and temperature. These highly turbulent flows typically enter the first stage vanes with significant spatial gradients near the inner and outer diameter endwalls. These gradients can result in secondary flow development within the vane passage that is different than what classical secondary flow models predict. The heat transfer between the working fluid and the turbine vane surface and endwalls is directly related to the secondary flows. The goal of the current study was to examine the migration of different inlet radial temperature and pressure Profiles through the high turbine vane of a modern turbine engine. The tests were performed using an inlet Profile Generator located in the Turbine Research Facility at the Air Force Research Laboratory. Comparisons of area-averaged radial exit Profiles are reported as well as Profiles at three vane pitch locations to document the circumferential variation in the Profiles. The results show that the shape of the total pressure Profile near the endwalls at the inlet of the vane can alter the redistribution of stagnation enthalpy through the airfoil passage significantly. Total pressure loss and exit flow angle variations are also examined for the different inlet Profiles.
-
Effects of Combustor Exit Profiles on Vane Aerodynamic Loading and Heat Transfer in a High Pressure Turbine
Journal of Turbomachinery, 2009Co-Authors: M. D. Barringer, Karen A. Thole, Marc D. PolankaAbstract:The flow and thermal fields exiting gas turbine combustors dictate the overall performance of the downstream turbine. The goal of this work was to investigate the effects of engine representative combustor exit Profiles on high pressure turbine vane aerodynamics and heat transfer. The various Profiles were produced using a nonreacting turbine inlet Profile Generator in the Turbine Research Facility (TRF) located at the Air Force Research Laboratory (AFRL). This paper reports how the pressure loading and heat transfer along the vane surface was affected by different turbine inlet pressure and temperature Profiles at different span locations. The results indicate that the inlet total pressure Profiles affected the aerodynamic loading by as much as 10%. The results also reveal that the combination of different total pressure and total temperature Profiles significantly affected the vane heat transfer relative to a baseline test with uniform inlet total pressure and total temperature. Near the inner diameter endwall, the baseline heat transfer was reduced 30‐40% over the majority of the vane surface. Near the outer dimeter endwall, it was found that certain inlet Profiles could increase the baseline heat transfer by 10‐ 20%, while other Profiles resulted in a decrease in the baseline heat transfer by 25‐35%. This study also shows the importance of knowing an accurate prediction of the local flow driving temperature when determining vane surface heat transfer. DOI: 10.1115/1.2950051
-
Migration of Combustor Exit Profiles Through High Pressure Turbine Vanes
Volume 4: Turbo Expo 2007 Parts A and B, 2007Co-Authors: M. D. Barringer, Marc D. Polanka, John P. Clark, P. J. Koch, Karen A. TholeAbstract:The high pressure turbine stage within gas turbine engines is exposed to combustor exit flows that are nonuniform in both stagnation pressure and temperature. These highly turbulent flows typically enter the first stage vanes with significant spatial gradients near the inner and outer diameter endwalls. These gradients can result in secondary flow development within the vane passage that is different than what classical secondary flow models predict. The heat transfer between the working fluid and the turbine vane surface and endwalls is directly related to the secondary flows. The goal of the current study was to examine the migration of different inlet radial temperature and pressure Profiles through the high turbine vane of a modern turbine engine. The tests were performed using an inlet Profile Generator located in the Turbine Research Facility (TRF) at the Air Force Research Laboratory (AFRL). Comparisons of area-averaged radial exit Profiles are reported as well as Profiles at three vane pitch locations to document the circumferential variation in the Profiles. The results show that the shape of the total pressure Profile near the endwalls at the inlet of the vane can alter the redistribution of stagnation enthalpy through the airfoil passage significantly. Total pressure loss and exit flow angle variations are also examined for the different inlet Profiles.Copyright © 2007 by ASME
-
Experimental Evaluation of an Inlet Profile Generator for High-Pressure Turbine Tests
Journal of Turbomachinery, 2006Co-Authors: M. D. Barringer, Karen A. Thole, Marc D. PolankaAbstract:Improving the performance and durability of gas turbine aircraft engines depends highly on achieving a better understanding of the flow interactions between the combustor and turbine sections. The flow exiting the combustor is very complex and it is characterized primarily by elevated turbulence and large variations in temperature and pressure. The heat transfer and aerodynamic losses that occur in the turbine passages are driven primarily by these spatial variations. To better understand these effects, the goal of this work is to benchmark an adjustable turbine inlet Profile Generator for the Turbine Research Facility (TRF) at the Air Force Research Laboratory. The research objective was to experimentally evaluate the performance of the nonreacting simulator that was designed to provide representative combustor exit Profiles to the inlet of the TRF turbine test section. This paper discusses the verification testing that was completed to benchmark the performance of the Generator. Results are presented in the form of temperature and pressure Profiles as well as turbulence intensity and length scale. This study shows how a single combustor geometry can produce significantly different flow and thermal field conditions entering the turbine. Engine designers should place emphasis on obtaining accurate knowledge of the flow distribution within the combustion chamber. Turbine inlet conditions with significantly different Profile shapes can result in altered flow physics that can change local aerodynamics and heat transfer.
Kengo Kinoshita - One of the best experts on this subject based on the ideXlab platform.
-
De novo Profile generation based on sequence context specificity with the long short-term memory network
BMC Bioinformatics, 2018Co-Authors: Kazunori D. Yamada, Kengo KinoshitaAbstract:Background Long short-term memory (LSTM) is one of the most attractive deep learning methods to learn time series or contexts of input data. Increasing studies, including biological sequence analyses in bioinformatics, utilize this architecture. Amino acid sequence Profiles are widely used for bioinformatics studies, such as sequence similarity searches, multiple alignments, and evolutionary analyses. Currently, many biological sequences are becoming available, and the rapidly increasing amount of sequence data emphasizes the importance of scalable Generators of amino acid sequence Profiles. Results We employed the LSTM network and developed a novel Profile Generator to construct Profiles without any assumptions, except for input sequence context. Our method could generate better Profiles than existing de novo Profile Generators, including CSBuild and RPS-BLAST, on the basis of Profile-sequence similarity search performance with linear calculation costs against input sequence size. In addition, we analyzed the effects of the memory power of LSTM and found that LSTM had high potential power to detect long-range interactions between amino acids, as in the case of beta-strand formation, which has been a difficult problem in protein bioinformatics using sequence information. Conclusion We demonstrated the importance of sequence context and the feasibility of LSTM on biological sequence analyses. Our results demonstrated the effectiveness of memories in LSTM and showed that our de novo Profile Generator, SPBuild, achieved higher performance than that of existing methods for Profile prediction of beta-strands, where long-range interactions of amino acids are important and are known to be difficult for the existing window-based prediction methods. Our findings will be useful for the development of other prediction methods related to biological sequences by machine learning methods.
-
De novo Profile generation based on sequence context specificity with the long short-term memory network
2017Co-Authors: Kazunori D. Yamada, Kengo KinoshitaAbstract:Long short-term memory (LSTM) is one of the most attractive deep learning methods to learn time series or contexts of input data. Increasing studies, including biological sequence analyses in bioinformatics, utilize this architecture. Amino acid sequence Profiles are widely used for bioinformatics studies, such as sequence similarity searches, multiple alignments, and evolutionary analyses. Currently, many biological sequences are becoming available, and the rapidly increasing amount of sequence data emphasizes the importance of scalable Generators of amino acid sequence Profiles. We employed the LSTM network and developed a novel Profile Generator to construct Profiles without any assumptions, except for input sequence context. Our method could generate better Profiles than existing de novo Profile Generators, including CSBuild and RPS-BLAST, on the basis of Profile-sequence similarity search performance with linear calculation costs against input sequence size. In addition, we analyzed the effects of the memory power of LSTM and found that LSTM had high potential power to detect long-range interactions between amino acids, as in the case of beta-strand formation, which has been a difficult problem in protein bioinformatics using sequence information. We demonstrated the importance of sequence context and the feasibility of LSTM on biological sequence analyses. Our results demonstrated the effectiveness of memories in LSTM and showed that our de novo Profile Generator, SPBuild, achieved higher performance than that of existing methods for Profile prediction of beta-strands, where long-range interactions of amino acids are important and are known to be difficult for the existing window-based prediction methods. Our findings will be useful for the development of other prediction methods related to biological sequences by machine learning methods.
Karen A. Thole - One of the best experts on this subject based on the ideXlab platform.
-
An Experimental Study of Combustor Exit Profile Shapes on Endwall Heat Transfer in High Pressure Turbine Vanes
Journal of Turbomachinery, 2009Co-Authors: M. D. Barringer, Karen A. Thole, Marc D. PolankaAbstract:The design and development of current and future gas turbine engines for aircraft propulsion have focused on operating the high pressure turbine at increasingly elevated temperatures and pressures. The drive toward thermal operating conditions near theoretical stoichiometric limits as well as increasingly stringent requirements on reducing harmful emissions both equate to the temperature Profiles exiting combustors and entering turbines becoming less peaked than in the past. This drive has placed emphasis on determining how different types of inlet temperature and pressure Profiles affect the first stage airfoil endwalls. The goal of the current study was to investigate how different radial Profiles of temperature and pressure affect the heat transfer along the vane endwall in a high pressure turbine. Testing was performed in the Turbine Research Facility located at the Air Force Research Laboratory using an inlet Profile Generator. Results indicate that the convection heat transfer coefficients are influenced by both the inlet pressure Profile shape and the location along the endwall. The heat transfer driving temperature for inlet Profiles that are nonuniform in temperature is also discussed.
-
Migration of Combustor Exit Profiles Through High Pressure Turbine Vanes
Journal of Turbomachinery, 2009Co-Authors: M. D. Barringer, Karen A. Thole, Marc D. Polanka, John P. Clark, P. J. KochAbstract:The high pressure turbine stage within gas turbine engines is exposed to combustor exit flows that are nonuniform in both stagnation pressure and temperature. These highly turbulent flows typically enter the first stage vanes with significant spatial gradients near the inner and outer diameter endwalls. These gradients can result in secondary flow development within the vane passage that is different than what classical secondary flow models predict. The heat transfer between the working fluid and the turbine vane surface and endwalls is directly related to the secondary flows. The goal of the current study was to examine the migration of different inlet radial temperature and pressure Profiles through the high turbine vane of a modern turbine engine. The tests were performed using an inlet Profile Generator located in the Turbine Research Facility at the Air Force Research Laboratory. Comparisons of area-averaged radial exit Profiles are reported as well as Profiles at three vane pitch locations to document the circumferential variation in the Profiles. The results show that the shape of the total pressure Profile near the endwalls at the inlet of the vane can alter the redistribution of stagnation enthalpy through the airfoil passage significantly. Total pressure loss and exit flow angle variations are also examined for the different inlet Profiles.
-
Effects of Combustor Exit Profiles on Vane Aerodynamic Loading and Heat Transfer in a High Pressure Turbine
Journal of Turbomachinery, 2009Co-Authors: M. D. Barringer, Karen A. Thole, Marc D. PolankaAbstract:The flow and thermal fields exiting gas turbine combustors dictate the overall performance of the downstream turbine. The goal of this work was to investigate the effects of engine representative combustor exit Profiles on high pressure turbine vane aerodynamics and heat transfer. The various Profiles were produced using a nonreacting turbine inlet Profile Generator in the Turbine Research Facility (TRF) located at the Air Force Research Laboratory (AFRL). This paper reports how the pressure loading and heat transfer along the vane surface was affected by different turbine inlet pressure and temperature Profiles at different span locations. The results indicate that the inlet total pressure Profiles affected the aerodynamic loading by as much as 10%. The results also reveal that the combination of different total pressure and total temperature Profiles significantly affected the vane heat transfer relative to a baseline test with uniform inlet total pressure and total temperature. Near the inner diameter endwall, the baseline heat transfer was reduced 30‐40% over the majority of the vane surface. Near the outer dimeter endwall, it was found that certain inlet Profiles could increase the baseline heat transfer by 10‐ 20%, while other Profiles resulted in a decrease in the baseline heat transfer by 25‐35%. This study also shows the importance of knowing an accurate prediction of the local flow driving temperature when determining vane surface heat transfer. DOI: 10.1115/1.2950051
-
Migration of Combustor Exit Profiles Through High Pressure Turbine Vanes
Volume 4: Turbo Expo 2007 Parts A and B, 2007Co-Authors: M. D. Barringer, Marc D. Polanka, John P. Clark, P. J. Koch, Karen A. TholeAbstract:The high pressure turbine stage within gas turbine engines is exposed to combustor exit flows that are nonuniform in both stagnation pressure and temperature. These highly turbulent flows typically enter the first stage vanes with significant spatial gradients near the inner and outer diameter endwalls. These gradients can result in secondary flow development within the vane passage that is different than what classical secondary flow models predict. The heat transfer between the working fluid and the turbine vane surface and endwalls is directly related to the secondary flows. The goal of the current study was to examine the migration of different inlet radial temperature and pressure Profiles through the high turbine vane of a modern turbine engine. The tests were performed using an inlet Profile Generator located in the Turbine Research Facility (TRF) at the Air Force Research Laboratory (AFRL). Comparisons of area-averaged radial exit Profiles are reported as well as Profiles at three vane pitch locations to document the circumferential variation in the Profiles. The results show that the shape of the total pressure Profile near the endwalls at the inlet of the vane can alter the redistribution of stagnation enthalpy through the airfoil passage significantly. Total pressure loss and exit flow angle variations are also examined for the different inlet Profiles.Copyright © 2007 by ASME
-
Experimental Evaluation of an Inlet Profile Generator for High-Pressure Turbine Tests
Journal of Turbomachinery, 2006Co-Authors: M. D. Barringer, Karen A. Thole, Marc D. PolankaAbstract:Improving the performance and durability of gas turbine aircraft engines depends highly on achieving a better understanding of the flow interactions between the combustor and turbine sections. The flow exiting the combustor is very complex and it is characterized primarily by elevated turbulence and large variations in temperature and pressure. The heat transfer and aerodynamic losses that occur in the turbine passages are driven primarily by these spatial variations. To better understand these effects, the goal of this work is to benchmark an adjustable turbine inlet Profile Generator for the Turbine Research Facility (TRF) at the Air Force Research Laboratory. The research objective was to experimentally evaluate the performance of the nonreacting simulator that was designed to provide representative combustor exit Profiles to the inlet of the TRF turbine test section. This paper discusses the verification testing that was completed to benchmark the performance of the Generator. Results are presented in the form of temperature and pressure Profiles as well as turbulence intensity and length scale. This study shows how a single combustor geometry can produce significantly different flow and thermal field conditions entering the turbine. Engine designers should place emphasis on obtaining accurate knowledge of the flow distribution within the combustion chamber. Turbine inlet conditions with significantly different Profile shapes can result in altered flow physics that can change local aerodynamics and heat transfer.
Kazunori D. Yamada - One of the best experts on this subject based on the ideXlab platform.
-
De novo Profile generation based on sequence context specificity with the long short-term memory network
BMC Bioinformatics, 2018Co-Authors: Kazunori D. Yamada, Kengo KinoshitaAbstract:Background Long short-term memory (LSTM) is one of the most attractive deep learning methods to learn time series or contexts of input data. Increasing studies, including biological sequence analyses in bioinformatics, utilize this architecture. Amino acid sequence Profiles are widely used for bioinformatics studies, such as sequence similarity searches, multiple alignments, and evolutionary analyses. Currently, many biological sequences are becoming available, and the rapidly increasing amount of sequence data emphasizes the importance of scalable Generators of amino acid sequence Profiles. Results We employed the LSTM network and developed a novel Profile Generator to construct Profiles without any assumptions, except for input sequence context. Our method could generate better Profiles than existing de novo Profile Generators, including CSBuild and RPS-BLAST, on the basis of Profile-sequence similarity search performance with linear calculation costs against input sequence size. In addition, we analyzed the effects of the memory power of LSTM and found that LSTM had high potential power to detect long-range interactions between amino acids, as in the case of beta-strand formation, which has been a difficult problem in protein bioinformatics using sequence information. Conclusion We demonstrated the importance of sequence context and the feasibility of LSTM on biological sequence analyses. Our results demonstrated the effectiveness of memories in LSTM and showed that our de novo Profile Generator, SPBuild, achieved higher performance than that of existing methods for Profile prediction of beta-strands, where long-range interactions of amino acids are important and are known to be difficult for the existing window-based prediction methods. Our findings will be useful for the development of other prediction methods related to biological sequences by machine learning methods.
-
De novo Profile generation based on sequence context specificity with the long short-term memory network
2017Co-Authors: Kazunori D. Yamada, Kengo KinoshitaAbstract:Long short-term memory (LSTM) is one of the most attractive deep learning methods to learn time series or contexts of input data. Increasing studies, including biological sequence analyses in bioinformatics, utilize this architecture. Amino acid sequence Profiles are widely used for bioinformatics studies, such as sequence similarity searches, multiple alignments, and evolutionary analyses. Currently, many biological sequences are becoming available, and the rapidly increasing amount of sequence data emphasizes the importance of scalable Generators of amino acid sequence Profiles. We employed the LSTM network and developed a novel Profile Generator to construct Profiles without any assumptions, except for input sequence context. Our method could generate better Profiles than existing de novo Profile Generators, including CSBuild and RPS-BLAST, on the basis of Profile-sequence similarity search performance with linear calculation costs against input sequence size. In addition, we analyzed the effects of the memory power of LSTM and found that LSTM had high potential power to detect long-range interactions between amino acids, as in the case of beta-strand formation, which has been a difficult problem in protein bioinformatics using sequence information. We demonstrated the importance of sequence context and the feasibility of LSTM on biological sequence analyses. Our results demonstrated the effectiveness of memories in LSTM and showed that our de novo Profile Generator, SPBuild, achieved higher performance than that of existing methods for Profile prediction of beta-strands, where long-range interactions of amino acids are important and are known to be difficult for the existing window-based prediction methods. Our findings will be useful for the development of other prediction methods related to biological sequences by machine learning methods.
