System and method for predicting binding affinity of small molecules to RNA targets

Abstract

The present disclosure is directed to a system 100 and a method 300 for predicting binding affinity of small molecules to Ribo-Nucleic Acid (RNA) targets The method comprises of receiving an input from a user, comprising a sequence of an RNA target, information related to a small molecule and a sub-type of the RNA target. The method further comprises of predicting a binding affinity of the small molecule to the RNA target of the sub-type, using a prediction model. Generating the prediction model comprises of, obtaining a plurality of RNA sequence based features for a plurality of sub-types of a plurality of RNA targets, and a plurality of small molecule features for a plurality of small molecules. Generating the prediction model further comprises of identifying a plurality of non-correlated features from the plurality of RNA sequence based features and the plurality of small molecule features and generating a final feature combination from the plurality of non- correlated features for each sub-type of the plurality of sub-types of the plurality of RNA targets based on a quasi-binding affinity of each of the plurality of small molecules with each sub-type of the plurality of sub-types of the plurality of RNA targets. The final feature combination is used to generate the prediction model.

Description

TITLE: SYSTEM AND METHOD FOR PREDICTING BINDING AFFINITY OF SMALL MOLECULES TO RNA TARGETSTECHNICAL FIELD:

[0001] The present disclosure generally relates computational biology and Ribo-Nucleic Acid (RNA)-small molecule interactions. More particularly, but not specifically, the present disclosure relates to a system and method for predicting binding affinity of small molecules to RNA targets.BACKGROUND:

[0002] Ribo-Nucleic Acid (RNA) is a central biomolecule in all cell types. The versatility of various RNA subtypes (mRNA, rRNA, tRNA, IncRNA, miRNA, snRNA and others) in terms of their functional profiles indicates their indispensable nature in cellulo. RNAs have acquired the ability to function flexibly in several biological pathways through their dynamic structure. They are involved in all stages of cellular regulation namely, pre-transcriptional, post-transcriptional and post-translational processes. Changes in RNA conformation can serve as a beacon to identify changes in cellular environment including, changes in cellular localization, availability of interaction partners, and altered metabolic state of a cell, indicating the sensitive and malleable nature of RNA structure. Therefore, detailed RNA structure profiling in both human cells and pathogens can enable better understanding of the alterations in cellular processes which can lead to disease conditions. The knowledge gathered from the structures of RNAs interacting with other biomolecules and their implications in disease-associated pathways can help delineate novel RNA targets for molecular therapy.

[0003] Several studies have experimentally proven the involvement of structured coding and noncoding RNA regions in a spectrum of diseases. Specifically, with increasing resistance to existing drugs targeting bacteria and viruses, and with the limited number of druggable target proteins identified in human body, it is the need of the hour to identify potential alternative therapeutic targets for drug development. Expanding the space of drug targets is essential in diseases such as cancers and genetic disorders wherein, targeted delivery of small molecular drugs is expensive owing to the inaccessible cellular localization of the protein targets involved. Complex and multifactorial diseases, such as cancers, might require simultaneous targeting of multiple regulatorybiomolecules to prevent the spread of the disease. This necessitates identification of drug targets participating in multiple levels of cellular, such as RNA molecules. With advancements in structure determination techniques, linear increase in the number of solved RNA structures has currently provided an impetus to several studies to probe the potential of RNAs to act as viable alternative therapeutic targets. Hence, it is essential to understand the mode of interaction of different RNA structural elements with drug-like small molecules, to elucidate the RNA-targeting chemical space and identify better selective inhibitors for structured RNAs involved in disease conditions.

[0004] In existing systems, Computer-Aided Drug Design (CADD) approaches have primarily been focused on the development of scoring functions or models, which can predict the in vitro activity of any given small molecule against a specific target protein or nucleic acid. Two such existing scoring methods from CADD and molecular modelling domains are, Quantitative Structure Activity Relationship (QSAR) models and molecular docking methods, respectively. While QSAR models can only predict the binding affinity of molecules toward specific targets, docking methods can provide a structure -level perspective of the target-ligand interaction, along with the predicted binding energy of each ligand pose. Existing systems also include Multi-Target (mt) QSAR models to account for activity against a family of targets rather than a single target, to address the problem of multi-targeted drug interventions. However, generic models specific to different RNA families are not available due to the limited experimental data available on binding affinities of RNA-small molecule interactions.

[0005] In view of one or more limitations stated above or any other limitations associated with the known arts and existing systems, there is a need for a generic-quantitative model to predict RNA- small molecule binding affinity.

[0006] The information disclosed in this background of the disclosure section is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person skilled in the art.SUMMARY:

[0007] It is an objective of the present disclosure to provide a system and method for predicting binding affinity of small molecules to Ribo-Nucleic Acid (RNA) targets.

[0008] To better address one or more of the previously mentioned concerns, in a first aspect of the disclosure there is provided a method for predicting binding affinity of small molecules to Ribo- Nucleic Acid (RNA) targets. The method comprises of receiving an input from a user, comprising a sequence of an RNA target, information related to a small molecule and a sub-type of the RNA target. The method further comprises of predicting a binding affinity of the small molecule to the RNA target of the sub-type, using a prediction model. Generating the prediction model comprises of, obtaining a plurality of RNA sequence based features for a plurality of sub-types of a plurality of RNA targets, and a plurality of small molecule features for a plurality of small molecules. Generating the prediction model further comprises of identifying a plurality of non-correlated features from the plurality of RNA sequence based features and the plurality of small molecule features and generating a final feature combination from the plurality of non-correlated features for each sub-type of the plurality of sub-types of the plurality of RNA targets based on a quasibinding affinity of each of the plurality of small molecules with each sub-type of the plurality of sub-types of the plurality of RNA targets. The final feature combination is used to generate the prediction model.

[0009] Further the present disclosure relates to a system for predicting binding affinity of small molecules to Ribo-Nucleic Acid (RNA) targets. The system comprises of a memory and a processor. The processor is configured to receive an input from a user, comprising a sequence of an RNA target, information related to a small molecule and a sub-type of the RNA target and predict a binding affinity of the small molecule to the RNA target of the sub-type, using a prediction model. The processor configured to generate the prediction model, is configured to, obtain a plurality of RNA sequence based features for a plurality of sub-types of a plurality of RNA targets, and a plurality of small molecule features for a plurality of small molecules. The processor is further configured to identify a plurality of non-correlated features from the plurality of RNA sequence based features and the plurality of small molecule features and generate a final feature combination from the plurality of non-correlated features for each sub-type of the plurality of sub-types of the plurality of RNA targets based on a quasi-binding affinity of each of theplurality of small molecules with each sub-type of the plurality of sub-types of the plurality of RNA targets. The final feature combination is used to generate the prediction model.

[0010] It is to be understood that the aspects and embodiments of the disclosure described above may be used in any combination with each other. Several of the aspects and embodiments may be combined together to form a further embodiment of the disclosure.

[0011] The foregoing summary is illustrative only and is not intended to be in any way limiting. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features will become apparent by reference to the drawings and the following detailed description.BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In the drawings, like reference characters generally refer to the same parts throughout the different views. Also, the drawings are not necessarily to scale, emphasis instead generally being placed upon illustrating the principles of the invention.

[0013] Fig. 1 discloses an environment for predicting binding affinity of a small molecule to an RNA target for implementing embodiments consistent with the present disclosure.

[0014] Fig. 2 is a detailed block diagram of a binding affinity predicting system for implementing embodiments consistent with the present disclosure.

[0015] Fig. 3 is a flowchart of a method for predicting binding affinity of small molecules to RNA targets, in accordance with an embodiment.

[0016] Fig. 4 is a flowchart of a method for generating the prediction model, in accordance with an embodiment.

[0017] Fig. 5 is a block diagram of an exemplary computing system for implementing embodiments consistent with the present disclosure.

[0018] Fig. 6 is an exemplary web based tool for implementing embodiments consistent with the present disclosure.

[0019] Fig. 7 depicts a first exemplary regression plot in accordance with an embodiment of the present disclosure.

[0020] Fig. 8 is a first exemplary interaction in accordance with an embodiment of the present disclosure.

[0021] Fig. 9 is a second exemplary interaction in accordance with an embodiment of the present disclosure.

[0022] Fig. 10 depicts a second exemplary regression plot in accordance with an embodiment of the present disclosure.DETAILED DESCRIPTION

[0023] In the present document, the word "exemplary" is used herein to mean "serving as an example, instance, or illustration." Any embodiment or implementation of the present subject matter described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.

[0024] While the disclosure is susceptible to various modifications and alternative forms, specific embodiment thereof has been shown by way of example in the drawings and will be described in detail below. It should be understood, however that it is not intended to limit the disclosure to the particular forms disclosed, but on the contrary, the disclosure is to cover all modifications, equivalents, and alternative falling within the spirit and the scope of the disclosure.

[0025] The terms “comprises”, “comprising”, or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a setup, device or method that comprises a list of components or steps does not include only those components or steps but may include other components or steps not expressly listed or inherent to such setup or device or method. In other words, one or more elements in a device or system or apparatus proceeded by “comprises... a” does not, without more constraints, preclude the existence of other elements or additional elements in the device or system or apparatus.

[0026] The terms "an embodiment", "embodiment", "embodiments", "the embodiment", "the embodiments", "one or more embodiments", "some embodiments", and "one embodiment" mean "one or more (but not all) embodiments of the invention(s)" unless expressly specified otherwise.

[0027] The terms "including", "comprising", “having” and variations thereof mean "including but not limited to", unless expressly specified otherwise.

[0028] As used herein, the terms “communication” and “communicate” may refer to the reception, receipt, transmission, transfer, provision, and / or the like of information (e.g., data, signals, messages, instructions, commands, and / or the like). For one unit (e.g., a device, a system, a component of a device or system, combinations thereof, and / or the like) to be in communication with another unit means that the one unit is able to directly or indirectly receive information from and / or transmit information to the other unit. This may refer to a direct or indirect connection (e.g., a direct communication connection, an indirect communication connection, and / or the like) that is wired and / or wireless in nature. Additionally, two units may be in communication with each other even though the information transmitted may be modified, processed, relayed, and / or routed between the first and second unit. For example, a first unit may be in communication with a second unit even though the first unit passively receives information and does not actively transmit information to the second unit. As another example, a first unit may be in communication with a second unit if at least one intermediary unit (e.g., a third unit located between the first unit and the second unit) processes information received from the first unit and communicates the processed information to the second unit. In some non-limiting embodiments, a message may refer to a network packet (e.g., a data packet and / or the like) that includes data. It will be appreciated that numerous other arrangements are possible.

[0029] As used herein, the term “computing system” may refer to one or more electronic devices or computers, that are configured to directly or indirectly communicate with or over one or more networks. A computing system may be a mobile or portable computing device, a desktop computer, a server, and / or the like. Furthermore, the term “computer” may refer to any computing system that includes the necessary components to receive, process, and output data, and normally includes a display, a processor, a memory, an input device, and a network interface.

[0030] As used herein, the term "user" may include an individual using the system for predicting binding affinity of small molecules to RNA targets.

[0031] As used herein, the term “server” or "server computer" is typically a powerful computer or cluster of computers. For example, the server computer can be a large mainframe, a minicomputer cluster, or a group of servers functioning as a unit. In one example, the server computer may be a database server coupled to a Web server.

[0032] As used herein, the term "processor" may refer to any suitable data computation device or devices. A processor may comprise one or more microprocessors working together to accomplish a desired function. The processor may include Central Processing Unit (CPU) comprises at least one high-speed data processor adequate to execute program components for executing user and / or system-generated requests. The CPU may be a microprocessor such as AMD's Athlon, Duron and / or Opteron; IBM and / or Motorola's PowerPC; IBM's and Sony's Cell processor; Intel's Celeron, Itanium, Pentium, Xeon, and / or XScale; and / or the like processor(s).

[0033] As used herein, the term "memory" may be any suitable device or devices that can store electronic data. A suitable memory may comprise a non-transitory computer readable medium that stores instructions that can be executed by a processor to implement a desired method. Examples of memories may comprise one or more memory chips, disk drives, etc. Such memories may operate using any suitable electrical, optical, and / or magnetic mode of operation.

[0034] As used herein, the term "database" may be an external database or internal database. The external database which may be publicly available. The internal database may be a database available within an organization and data available from regulatory agencies. The database maybe a computer system or in a server.

[0035] As used herein, the term “small molecule” or “small molecules” may be a low molecular weight (< 1000 Daltons) organic compound that may regulate a biological process. It may have a size on the order of 1 nm.

[0036] It will be apparent that systems and / or methods, described herein, can be implemented in different forms of hardware, software, or a combination of hardware and software. The actual specialized control hardware or software code used to implement these systems and / or methods is not limiting of the implementations. Thus, the operation and behaviour of the systems and / or methods are described herein without reference to specific software code, it being understood that software and hardware can be designed to implement the systems and / or methods based on the description herein.

[0037] Fig. 1 discloses an environment for predicting binding affinity of a small molecule to an RNA target for implementing embodiments consistent with the present disclosure.

[0038] In an embodiment, the binding affinity predicting system 100 maybe in direct connection or in connection via a network interface with a server 103 that may compute and store a plurality of RNA sequence based features data which may be acquired and stored in the memory 202.

[0039] In an embodiment, the binding affinity predicting system 100 may also be in direct connection or in connection via the network interface with one or more databases 104i, 1042, ... 104nthat may each store a plurality of small molecule-RNA target binding affinity values, which may be acquired and stored in the memory 202. Such databases may be implemented as fault- tolerant, relational, scalable, secure databases such as Oracle or Sybase.

[0040] In an embodiment, the network interface may employ connection protocols including, without limitation, direct connect, Ethernet (e.g., twisted pair 10 / 100 / 1000 Base T), Transmission Control Protocol / Internet Protocol (TCP / IP), token ring, IEEE 802.11 a / b / g / n / x, etc.

[0041] In an embodiment, the binding affinity predicting system 100 may be configured to predict binding affinity for a given small molecule with a given RNA target. In an embodiment, the binding affinity predicting system 100 may receive RNA sequence, small molecule and RNA sub-type as inputs. The binding affinity predicting system 100 may implement an Al model to predict the binding affinity of the small molecule with the RNA target.

[0042] Fig. 2 is a detailed block diagram of the binding affinity predicting system 100 for implementing embodiments consistent with the present disclosure.

[0043] In an embodiment, the binding affinity predicting system 100 may include, without limitation to, the processor 201, the memory 202, an I / O (Input / Output) interface 215, the network interface 216. The processor 201 may be configured to perform one or more functions of the binding affinity predicting system 100 using the data 214 and the one or more modules 206 stored in the memory 202. In an embodiment, the memory 202 may store the data 214 and the one or more modules 206.

[0044] In an embodiment, the data 214 stored in the memory 202 may include, without limitation, small molecule-RNA target binding affinity value 203, RNA sequence based features data 204 and other data 205. In an embodiment, the data 214 may be stored within the memory 202 in the form of various data structures. Additionally, the data 214 may be organized using data models, such as relational or hierarchical data models. The other data 205 may include various temporary data and files generated by the one or more modules 206.

[0045] In an embodiment, the small molecule-RNA target binding affinity value 203 may be data acquired by the binding affinity predicting system 100, and stored in the memory 202, from the one or more databases 104i, 1042, ... 104n. The small molecule-RNA target binding affinity value 203 may be a plurality of binding affinity values of a plurality of small molecules to a plurality of RNA targets. In an embodiment, the binding affinity values may be quantified in terms of, without limitations to, dissociation constant (Kd), association constant etc.

[0046] In an embodiment, to acquire the small molecule-RNA target binding affinity value 203, the binding affinity predicting system 100 may be in direct connection or in connection via the network interface 216 with one or more databases 104i, 1042, ... 104nthat may each store aplurality of small molecule-RNA target binding affinity values. Such databases may be implemented as fault-tolerant, relational, scalable, secure databases such as Oracle or Sybase.

[0047] In an embodiment, the RNA sequence based features data 204 may be data acquired by the binding affinity predicting system 100, and stored in the memory 202, from the server 103. The RNA sequence based features data 204 may be classified based on, without limitation to, three categories, oligonucleotide or K-tuple nucleotide composition, pseudo-nucleotide composition and structure composition.

[0048] In an embodiment, to acquire the RNA sequence based features 204, the binding affinity predicting system 100 may be in direct connection or in connection via the network interface 216 with the server 103 that may store the RNA sequence based features data 204. In an embodiment the RNA sequence based features 204 may be classified based on, without limitation to, three categories, oligonucleotide or K-tuple nucleotide composition, pseudo-nucleotide composition and structure composition.

[0049] In an embodiment, the network interface 216 may employ connection protocols including, without limitation, direct connect, Ethernet (e.g., twisted pair 10 / 100 / 1000 Base T), Transmission Control Protocol / Internet Protocol (TCP / IP), token ring, IEEE 802.11 a / b / g / n / x, etc.

[0050] In an embodiment, the data 214 may be processed by the one or more modules 206 of the binding affinity predicting system 100. In some implementations, the one or more modules 206 may be communicatively coupled to the processor 201 for performing one or more functions of the binding affinity predicting system 100. In an implementation, the one or more modules 206 may include, without limiting to, a receiving module 207, an RNA feature calculation module 208, a small molecule feature calculation module 209, a prediction model generating module 210, a prediction model 211, an executing module 212 and other modules 213.

[0051] As used herein, the term module may refer to an Application Specific Integrated Circuit (ASIC), an electronic circuit, a hardware processor (shared, dedicated, or group) and memory that executes one or more software or firmware programs, a combinational logic circuit, and / or other suitable components that provide the described functionality. As used herein, the term module may also refer to a software program that may be implemented by the processor 201. In animplementation, each of the one or more modules 206 may be configured as stand-alone hardware computing units. In an embodiment, the other modules 213 may be used to perform various miscellaneous functionalities of the binding affinity predicting system 100. It will be appreciated that one or more such modules 206 may be represented as a single module or a combination of different modules.

[0052] In an embodiment, the RNA feature calculation module 208 may be configured to receive the RNA sequence based features data 204. It may also be configured to store and run feature calculation scripts that compute a plurality of RNA sequence based features from the plurality of RNA target sequences from the one or more databases 104i, 1042, ... 104n.

[0053] In an embodiment, the small molecule feature calculation module 209 may be configured to calculate ID (Dimensional), 2D and 3D small molecule features from the plurality of small molecules data from the one or more databases 104i, 1042, ... 104n. The small molecule feature calculation module 209 may be configured to convert small molecules in Simplified Molecular Input Line Entry System (SMILES- ID) format to Structured Data files (SDF- 3D) format to enable 3D feature calculation.

[0054] In an embodiment, the receiving module 207 may be configured to receive the small molecule-RNA target binding affinity value 203 and may perform one or more pre-processing operations on the small molecule-RNA target binding affinity value 203. The one or more preprocessing operations may include, without limitation to, converting the small molecule-RNA target binding affinity values into Molar units, converting the small molecule-RNA target binding affinity values, which are in terms of Molar units, to log-scale to normalize the different concentration ranges that may exist in the small molecule-RNA target binding affinity value 203 and stratifying the log-scale small molecule-RNA target binding affinity values into plurality of sub-types of RNA. The plurality of sub-types of RNA may be, without limitation to, aptamers, miRNA, repeats, ribosomal RNA, riboswitches and viral RNA.

[0055] In an embodiment, the prediction model generating module 210 may be configured to generate the prediction model 211 which may be used to predict binding affinity of a small molecule to an RNA target. The prediction model generating module 210 may be configured toobtain a plurality of RNA sequence based features for a plurality of sub-types of a plurality of RNA targets from the RNA feature calculation module 208. The prediction model generating module 210 may be configured to obtain a plurality of small molecule features for a plurality of small molecules from the small molecule feature calculation module 209.

[0056] In an embodiment, the prediction model generating module 210 may be configured to identify a plurality of non-correlated features from the plurality of RNA sequence based features and the plurality of small molecule features. The plurality of non-correlated features may be identified using statistical methods, such as, without limitation to, Forward Feature Selection (FFS) method.

[0057] In an embodiment, the prediction model generating module 210 may be configured to generate a final feature combination from the plurality of non-correlated features for each sub-type of the plurality of sub-types of the plurality of RNA targets based on a quasi-binding affinity of each of the plurality of small molecules with each sub-type of the plurality of sub-types of the plurality of RNA targets.

[0058] In an embodiment, the prediction model generating module 210 may be configured to use the final feature combination to generate the prediction model 211.

[0059] In an embodiment, the prediction model generating module 210, may be further configured to identify the final feature by performing an iterative process may be repeated until the final feature combination is identified. The prediction model generating module 210 may be configured to iteratively input a feature combination of the plurality of small molecule features and the plurality of RNA sequence based features for each sub-type, to the prediction model 211.The prediction model generating module 210 may input a plurality of small molecule-RNA target pairs for a plurality of sub types of RNA, to the prediction model 211 and compute respective plurality of quasi-binding affinities. Thereafter, the prediction model generating module 210 may be configured to determine one or more performance variables for the plurality of quasi-binding affinities of the plurality of small molecule-RNA target pairs for the plurality of sub types of RNA. The one or more performance variables may be, without limitation, Pearson correlation coefficient (r) and Mean Absolute Error (MAE). Pearson correlation coefficient may be calculated by:MAE may be calculated by:Where, n may be the total number of small molecule-RNA target pairs input into the prediction model 211, i is the 1thsmall molecule-RNA target pair, Xi is the predicted value of binding affinity for the small molecule-RNA target pair, and yi is the observed value of the small molecule-RNA target obtained from the log-scale small molecule-RNA target binding affinity values obtained from the receiving module 207.In an embodiment, the feature combination of an iteration may be identified as final feature combination if the values of the one or more performance variables are within respective threshold range. The threshold range for ‘r’ may be, without limitation, for example ‘r’ greater than 0.6 . The threshold range for MAE may be, without limitation, for example MAE lesser than 1.0.In an embodiment, if the values of the one or more performance variables are outside respective threshold range, the prediction model generating module 210, may increment the number of features in the feature combination and re-iterate the above described process.In an embodiment, the prediction model generating module 210 may generate the prediction model 211 to predict the binding affinity using the identified final feature combination.

[0060] In an embodiment, all possible features combinations, of the plurality of small molecule features and the plurality of RNA sequence based features for each sub-type, for a particular number of features may be considered for each iteration of the iterative process. All the possible feature combinations are input into the prediction model 211. If the one or more performance variables for a feature combination of one or more small molecules and one or more RNA targets, for a sub-type of RNA, are outside respective threshold range, that feature combination may be omitted. For example, without limitation, in an iteration, the plurality of small molecule features and the plurality of RNA sequence based features for each sub-type of RNA, are being considered to form feature combinations, with three features in each combination. If the one or moreperformance variables of one of the feature combinations falls outside respective threshold range, that feature combination may be omitted. Subsequently, one or more performance variables may be calculated for all the other possible three feature combinations, after the omission of one of the feature combinations. After, all the possible three feature combinations are run through the prediction model, then the number of features per feature combination may be incremented, i.e., the number of features per feature combination may be incremented to four features per combination and the whole process may be re-iterated with the unique set of features obtained from the successful three feature combinations.

[0061] In an embodiment, the prediction model generating module 210, may identify the final feature combination as the feature combination with the smallest number of features, offering optimal performance (performance is measured in terms of the one or more performance variables). For e.g., without limitation, if a feature combination comprising of three features and another feature combination comprising of four features, both result in similar performance variables, then the feature combination comprising of three features may be considered as the final feature combination. This has the advantageous effect of the performance of the binding affinity predicting system 100 being improved and also a reduction in the feature calculation load for the binding affinity predicting system 100.

[0062] In an embodiment, the executing module 212 may be configured to execute the prediction model 211.

[0063] Fig. 3 depicts a flowchart of a method 300 for predicting binding affinity of small molecules to RNA targets, in accordance with an embodiment.

[0064] As illustrated in Fig. 3, the method 300 may include one or more blocks illustrating a method for predicting binding affinity of small molecules to RNA targets using the binding affinity predicting system 100 illustrated in Fig. 2. The method 300 may be described in the general context of computer executable instructions. Generally, computer executable instructions can include routines, programs, objects, components, data structures, procedures, modules, and functions, which perform specific functions or implement specific abstract data types.

[0065] The order in which the method 300 is described is not intended to be construed as a limitation, and any number of the described method blocks can be combined in any order to implement the method. Additionally, individual blocks may be deleted from the methods without departing from the scope of the subject matter described herein. Furthermore, the method can be implemented in any suitable hardware, software, firmware, or combination thereof.

[0066] At block 301, the method 300 comprises of receiving, by the processor 201 of the binding affinity predicting system 100 for predicting binding affinity of small molecules to RNA targets, an input from a user, comprising a sequence of an RNA target, information related to a small molecule and a sub-type of the RNA target. The sub-types of the RNA target may be one of aptamers, miRNA, repeats, ribosomal RNA, riboswitches or viral RNA. At block 302, the method 300 comprises of predicting, by the processor 201, a binding affinity of the small molecule to the RNA target of the sub-type, using the prediction model 211.

[0067] At block 303, of the method 300, generating the prediction model 211 comprises of obtaining at block 304 of the method 300, by the processor 201 , a plurality of RNA sequence based features for a plurality of sub-types of a plurality of RNA targets, and a plurality of small molecule features for a plurality of small molecules. The plurality of RNA sequence based features for the plurality of sub-types of the plurality of RNA targets may be adapted from the server 103 and may be classified based on, without limitation to, three categories, oligonucleotide or K-tuple nucleotide composition, pseudo-nucleotide composition and structure composition. The plurality of RNA sequence based features may also be calculated by the RNA feature calculation module 208 of the binding affinity predicting system 100. The plurality of small molecule features may be calculated by the small molecule feature calculation module 209 of the binding affinity predicting system 100.

[0068] At block 305, the method 300 comprises of identifying, by the processor 201, a plurality of non-correlated features from the plurality of RNA sequence based features and the plurality of small molecule features. The plurality of non-correlated features may be identified using statistical methods, such as, without limitation to, Forward Feature Selection (FFS) method.

[0069] At block 306, the method 300 comprises of generating, by the processor 201 , a final feature combination from the plurality of non-correlated features for each sub-type of the plurality of subtypes of the plurality of RNA targets based on a quasi-binding affinity of each of the plurality of small molecules with each sub-type of the plurality of sub-types of the plurality of RNA targets, wherein the final feature combination is used to generate the prediction model 211.

[0070] Fig. 4 is a flowchart of a method 400 for generating the prediction model 211 , in accordance with an embodiment.

[0071] As illustrated in Fig. 4, the method 400 may include one or more blocks illustrating a method for generating the prediction model 211 using the binding affinity predicting system 100 illustrated in Fig. 2. The method 400 may be described in the general context of computer executable instructions. Generally, computer executable instructions can include routines, programs, objects, components, data structures, procedures, modules, and functions, which perform specific functions or implement specific abstract data types.

[0072] The order in which the method 400 is described is not intended to be construed as a limitation, and any number of the described method blocks can be combined in any order to implement the method. Additionally, individual blocks may be deleted from the methods without departing from the scope of the subject matter described herein. Furthermore, the method can be implemented in any suitable hardware, software, firmware, or combination thereof.

[0073] Further to block 303 of the method 300, generating the prediction model 211 further comprises, iteratively performing, until identifying the final feature combination, the following blocks of method 400.

[0074] At block 401, the method 400 comprises of inputting, by the processor 201, a feature combination of the plurality of small molecule features and the plurality of RNA sequence based features for each sub-type, to the prediction model 211 i.e., a feature combination of the plurality of small molecule features and the plurality of RNA sequence based features for each sub-type, to the prediction model 211.

[0075] At block 402, the method 400 comprises of, determining, by the processor 201, the quasibinding affinity of the plurality of small molecules with each sub-type of the plurality of sub-types of the plurality of RNA targets i.e., the processor 201 may input a plurality of small molecule- RNA target pairs for a plurality of sub types of RNA, to the prediction model 211 and compute respective plurality of quasi-binding affinities.

[0076] At block 403, the method 400 comprises of, determining, by the processor 201 , one or more performance variables for the quasi-binding affinity of the plurality of small molecules with each sub-type of the plurality of sub-types of the plurality of RNA targets, i.e., the processor 201 may be configured to determine one or more performance variables for the plurality of quasi-binding affinities of the plurality of small molecule-RNA target pairs for the plurality of sub types of RNA. The one or more performance variables may be, without limitation, Pearson correlation coefficient (r) and Mean Absolute Error (MAE). Pearson correlation coefficient may be calculated by:MAE may be calculated by:Where, n may be the total number of small molecule-RNA target pairs input into the prediction model 211, i is the 1thsmall molecule-RNA target pair, Xi is the predicted value of binding affinity for the small molecule-RNA target pair, and yi is the observed value of the small molecule-RNA target obtained from the log-scale small molecule-RNA target binding affinity values obtained from the receiving module 207.

[0077] At block 404, the method 400 comprises of identifying, by the processor 201, the feature combination as the final feature combination when values of the one or more performance variables are within respective threshold range, i.e., the feature combination of an iteration may be identified as final feature combination if the values of the one or more performance variables are within respective threshold range. The threshold range for ‘r’ may be, without limitation, for example ‘r’ greater than 0.6 . The threshold range for MAE may be, without limitation, for example MAE lesser than 1.0.

[0078] At block 405, the method 400 comprises of, Incrementing number of features in the feature combination when the values of the one or more performance variables are outside respective threshold range i.e., if the values of the one or more performance variables are outside respective threshold range, the processor 201, may increment the number of features in the feature combination and re-iterate the above described blocks.

[0079] At block 406, the method 400 comprises of, generating, by the processor 201, the prediction model 211 to predict the binding affinity using the final feature combination.

[0080] In an embodiment, the processor 201, may identify the final feature combination as the feature combination with the smallest number of features, offering optimal performance (performance is measured in terms of the one or more performance variables). For e.g., without limitation, if a feature combination comprising of three features and another feature combination comprising of four features, both result in similar performance variables, them the feature combination comprising of three features may be considered as the final feature combination.

[0081] Fig. 5 is a block diagram of an exemplary computing system 500 for implementing embodiments consistent with the present disclosure.

[0082] In some embodiments, the computing system 500 may be the system 100 for predicting binding affinity of small molecules to RNA targets. The processor 201 may include at least one data processor for executing program components for executing user or system-generated business processes. A user may include a person, a person using a device such as those included in this disclosure, or such a device itself. The processor 201 may include specialized processing units such as integrated system (bus) controllers, memory management control units, floating point units, graphics processing units, digital signal processing units, etc.

[0083] The processor 201 may be disposed in communication with input devices 508 and output devices 509 via I / O interface 215. Input devices 508 may be, without limitation, a mouse, a keyboard, a microphone a bio-metric authentication device. Output devices 509 may be, without limitation, a monitor and a speaker. The VO interface 215 may employ communication protocols / methods such as, without limitation, audio, analog, digital, stereo, IEEE- 1393, serial bus,Universal Serial Bus (USB), infrared, PS / 2, BNC, coaxial, component, composite, Digital Visual Interface (DVI), High-Definition Multimedia Interface (HDMI), Radio Frequency (RF) antennas, S-Video, Video Graphics Array (VGA), IEEE 8O2.n / b / g / n / x, Bluetooth, cellular (e.g., Code- Division Multiple Access (CDMA), High-Speed Packet Access (HSPA+), Global System For Mobile Communications (GSM), Long-Term Evolution (LTE), WiMax, or the like), etc.

[0084] Using the I / O interface 215, the computing system 500 may communicate with the input devices 508 and the output devices 509.

[0085] In some embodiments, the processor 201 may be disposed in communication with a communication network 510 via the network interface 216. The network interface 216 may communicate with the communication network 510. The network interface 216 may employ connection protocols including, without limitation, direct connect, Ethernet (e.g., twisted pair 10 / 100 / 1000 Base T), Transmission Control Protocol / Internet Protocol (TCP / IP), token ring, IEEE 802.1 la / b / g / n / x, etc. In an embodiment, using the network interface 216 and the communication network 510, the computing system 500 may communicate with a plurality of databases 104i, 1042, ... 104nwhich may each store a plurality of small molecule-RNA target binding affinity values. Such databases may be implemented as fault-tolerant, relational, scalable, secure databases such as Oracle or Sybase. In an embodiment, using the network interface 216 and the communication network 510, the computing system 500 may communicate with the server 103 that may store the plurality of RNA sequence based features data.

[0086] The communication network 510 can be implemented as one of the different types of networks, such as intranet or Local Area Network (LAN), Closed Area Network (CAN) and such. The communication network 510 may either be a dedicated network or a shared network, which represents an association of the different types of networks that use a variety of protocols, for example, Hypertext Transfer Protocol (HTTP), CAN Protocol, Transmission Control Protocol / Internet Protocol (TCP / IP), Wireless Application Protocol (WAP), etc., to communicate with each other. Further, the communication network 510 may include a variety of network devices, including routers, bridges, servers, computing devices, storage devices, etc. In some embodiments, the processor 201 may be disposed in communication with the memory 202 (e.g.,RAM, ROM, etc. not shown in Fig. 5) via a storage interface 503. The storage interface 503 may connect to the memory 202 including, without limitation, memory drives, removable disc drives, etc., employing connection protocols such as Serial Advanced Technology Attachment (SATA), Integrated Drive Electronics (IDE), IEEE- 1393, Universal Serial Bus (USB), fibre channel, Small Computer Systems Interface (SCSI), etc. The memory drives may further include a drum, magnetic disc drive, magneto-optical drive, optical drive, Redundant Array of Independent Discs (RAID), solid-state memory devices, solid-state drives, etc.

[0087] The memory 202 may store a collection of program or database components, including, without limitation, a user interface 504, an operating system 505, a web browser 506, a mobile application 507 etc.

[0088] The operating system 505 may facilitate resource management and operation of the computing system 500. Examples of operating systems include, without limitation, APPLE® MACINTOSH® OS X®, UNIX®, UNIX-like system distributions (e.g.., BERKELEY SOFTWARE DISTRIBUTION® (BSD), FREEBSD®, NETBSD®, OPENBSD, etc.), LINUX® DISTRIBUTIONS (E.G., RED HAT®, UBUNTU®, KUBUNTU®, etc.), IBM®OS / 2®, MICROSOFT® WINDOWS® (XP®, VISTA® / 7 / 8, 10 etc.), APPLE® IOS®, GOOGLE™ ANDROID™, BLACKBERRY® OS, or the like. The User interface 504 may facilitate display, execution, interaction, manipulation, or operation of program components through textual or graphical facilities. For example, user interfaces may provide computer interaction interface elements on a display system operatively connected to the computing system 500, such as cursors, icons, checkboxes, menus, scrollers, windows, widgets, etc. Graphical User Interfaces (GUIs) may be employed, including, without limitation, Apple® Macintosh® operating systems’ Aqua®, IBM® OS / 2®, Microsoft® Windows® (e.g., Aero, Metro, etc.), web interface libraries (e.g., ActiveX®, Java®, Javascript®, AJAX, HTML, Adobe® Flash®, etc.), or the like.

[0089] In some embodiments, the computing system 500 may implement the web browser 506 stored program components. The web browser 506 may be a hypertext viewing application, such as MICROSOFT® INTERNET EXPLORER®, GOOGLE™ CHROME™, MOZILLA® FIREFOX®, APPLE® SAFARI®, etc. Secure web browsing may be provided using SecureHypertext Transport Protocol (HTTPS), Secure Sockets Layer (SSL), Transport Layer Security (TLS), etc. Web browsers 506 may utilize facilities such as AJAX, DHTML, ADOBE® FLASH®, JAVASCRIPT®, JAVA®, Application Programming Interfaces (APIs), etc. In some embodiments, the computing system 500 may implement a mail server stored program component. The mail server may be an Internet mail server such as Microsoft Exchange, or the like. The mail server may utilize facilities such as Active Server Pages (ASP), ACTIVEX®, ANSI® C++ / C#, MICROSOFT®, .NET, CGI SCRIPTS, JAVA®, JAVASCRIPT®, PERL®, PHP, PYTHON®, WEBOBJECTS®, etc. The mail server may utilize communication protocols such as Internet Message Access Protocol (IMAP), Messaging Application Programming Interface (MAPI), MICROSOFT® exchange, Post Office Protocol (POP), Simple Mail Transfer Protocol (SMTP), or the like. In some embodiments, the computing system 500 may implement a mail client stored program component. The mail client may be a mail viewing application, such as APPLE® MAIL, MICROSOFT® ENTOURAGE®, MICROSOFT® OUTLOOK®, MOZILLA® THUNDERBIRD®, etc.

[0090] Furthermore, one or more computer-readable storage media may be utilized in implementing embodiments consistent with the present disclosure. A computer-readable storage medium refers to any type of physical memory on which information or data readable by a processor may be stored. Thus, a computer-readable storage medium may store instructions for execution by one or more processors, including instructions for causing the processor(s) to perform steps or stages consistent with the embodiments described herein. The term “computer-readable medium” should be understood to include tangible items and exclude carrier waves and transient signals, i.e., non-transitory. Examples include Random Access Memory (RAM), Read-Only Memory (ROM), volatile memory, non-volatile memory, hard drives, Compact Disc (CD) ROMs, Digital Video Disc (DVDs), flash drives, disks, and any other known physical storage media.IMPLEMENTATION

[0091] Table 1 summarizes the performance comparison between the prediction model 211 and an existing QSAR model.

[0092] Further, another model specific to a subtype of RNA (Human Immuno -deficiency Virus Trans-Activation Response (HIV TAR RNA)-specific) is developed and considered for testing against the existing models.

[0093] Table 1 : Performance of the Viral RNA models on the test dataset curated from an existing QSAR study. Performance is measured using Pearson’s correlation coefficient (r) and Mean Absolute Error (MAE). The footnote indicates the difference in test datasets between the models provided."Test set consists of all 48 compounds from the QSAR study; *Test set consists of the 12 compounds from the test set of the QSAR study.

[0094] Based on the performance of the viral RNA models on a test dataset, the model specific to HIV TAR RNA (r = 0.89; MAE = 0.41) outperforms the model reported in the existing QSAR study (r = 0.86; MAE = 0.93) in terms of the both the performance metrics considered. Theregression plots comparing both the viral RNA model predictions (Generic and HIV TAR RNA- specific) with that of the model from the existing QSAR study are disclosed in Fig. 7.

[0095] Fig. 7 depicts a first exemplary regression plot comparing the predictions for the 12 compounds from the test dataset of the HIV QSAR study with that of the HIV-1 TAR NA-specific viral RNA model.

[0096] The performance of the feature combinations identified for each RNA subtype in predicting RNA-small molecule binding affinity is evaluated using stratified 10-fold cross-validation (10- fold CV) and jack-knife test (LOO-CV). The training, cross-validation and jack-knife metrics for all the models are summarized in Table 2.Table 2: Pearson correlation (r) and mean absolute error (MAE) for each binding affinity prediction model during training, stratified 10-fold CV, and jack-knife test.

[0097] The number of features involved in prediction, range from 8-13, since the feature selection procedure tries to identify the minimal feature combination with best predictive performance. The number of features disclosed herein must not be construed as a limitation. The mean correlation of all models is found to be 0.82632 and the mean MAE is 0.65839 in the jack-knife test.

[0098] Fig. 10 depicts a second exemplary regression plot comparing training and jack-knife test predictions with the experimental binding affinity values for each RNA subtype. Fig. 10 depicts regression plots showing the agreement between training and jack-knife test predictions for the binding affinity prediction models specific to (a) Aptamers, (b) miRNAs, (c) Repeats, (d)Ribosomal RNAs, (e) Riboswitches, and (f) Viral RNAs. In Fig. 10, the 45° diagonal line is provided as a black solid line and the ±1 error margins are provided as dotted lines.

[0099] The performance of models were evaluated with five external blind test datasets - four classification tasks and one regression task. The predicted binding affinity values from regression models were converted into classification results by using a threshold value of 4.0 to define actives and inactives for a given RNA target sequence. This threshold is selected based on the ligand concentration utilized in the small-molecular microarray (SMM) experiments used to curate the classification dataset. Area Under Receiver Operating Characteristic curve (AUROC) score, Flscore, sensitivity / recall, specificity, average precision, and Matthew’s Correlation Coefficient (MCC) were used as the metrics to evaluate classification performance and have been summarized in Table 3 below. It may be ascertained that the models can consistently classify molecules into actives and inactives with high sensitivity and specificity for the threshold of choice.

[0100] Table 3 depicts performance of the binding affinity prediction models on four classification datasets.

[0101] The performance of the viral RNA model on regression task is evaluated for 48 compounds. Based on the results, the viral RNA model is refined further by augmenting the training dataset with the training dataset, and evaluated on the 12 compounds which were part of the test dataset of the existing QSAR study.

[0102] As disclosed in Fig. 6 the binding affinity predicting system 100 has been implemented as a web server tool named RSAPred. The web server may take an RNA sequence in FASTA format and a small molecule in SMILES (ID) or SDF (3D) format as inputs, along with a user-preferred RNA subtype for binding affinity prediction, and a log-scale binding affinity (pKd) of the input RNA sequence and the input small molecule will be displayed to the user.

[0103] Two exemplary scenarios, illustrating the use of RSAPred to predict the binding affinity of RNA-small molecule interactions are depicted in Fig. 8 and Fig. 9. The exemplary scenarios must not be construed as a limitation. The RNA subtype of interest in the first exemplary scenario (as disclosed in Fig. 8) is a biotin-binding RNA aptamer (Protein Data Bank (PDB) ID: 1F27; RNA-Small molecule Interaction Miner (R-SIM) entry: 288) belonging to the aptamers subtype of RNA targets. Upon substitution of the feature values in the prediction model, the predicted pKd value for biotin is 5.10, which is very close to the experimentally measured binding affinity value (5.22). Based on the pKd value, it is predicted that biotin will have an effective concentration in the millimolar range (mM).Fig. 8 discloses a three-dimensional interaction observed between biotin and the biotin-binding RNA aptamer (PDB ID: 1F27; R-SIM entry: 288). The predominant interactions observed include magnesium ion chelation and hydrogen bonds (yellow dashes). The ligand is shown in green sticks, binding site nucleotides are shown in lines, magnesium ion is shown as a green sphere, and water molecules are shown as red spheres.

[0104] The RNA subtype in the second exemplary scenario (as disclosed in Fig. 9) is an acceptor site (A-site) of the E.coli 16S ribosomal RNA belonging to the ribosomal RNA subtype of RNA targets. Fig. 9 discloses a three-dimensional interaction observed between kanamycin and the 16S rRNA A-site (PDB ID: 2ESI; R-SIM entry: 599). The ligand is shown in green sticks, binding site nucleotides are shown in lines, and water molecule is shown in red sphere. The results from RSAPred show that the predicted pKd value for kanamycin is 5.25, which is very close to the experimentally measured binding affinity value (5.6). Based on the pKd value, it is predicted that kanamycin will have an effective concentration in the millimolar range (mM).

[0105] The selection of non-correlated features based on the as described method has the advantageous that the effect that the least number of features, giving optimal performance maybe selected for developing the prediction models. Further, this also has reduces the load of featurecalculation for the system. The system 100 for predicting binding affinity of small molecules to RNA targets helps predict the binding of affinity of novel small molecules to RNA targets.

[0106] Finally, the language used in the specification has been principally selected for readability and instructional purposes, and it may not have been selected to delineate or circumscribe the inventive subject matter. Accordingly, the disclosure of the embodiments of the disclosure is intended to be illustrative, but not limiting, of the scope of the disclosure.

[0107] With respect to the use of substantially any plural and / or singular terms herein, those having skill in the art can translate from the plural to the singular and / or from the singular to the plural as is appropriate to the context and / or application. The various singular / plural permutations may be expressly set forth herein for sake of clarity.

[0108] Any of the software components or functions described in this application, may be implemented as software code to be executed by a processor using any suitable computer language such as, for example, Java, C++ or Perl using, for example, conventional or object- oriented techniques. The software code may be stored as a series of instructions, or commands on a computer readable medium, such as a random access memory (RAM), a read only memory (ROM), a magnetic medium such as a hard-drive or a floppy disk, or an optical medium such as a CD-ROM. Any such computer readable medium may reside on or within a single computational apparatus, and may be present on or within different computational apparatuses within a system or network.

[0109] The above description is illustrative and is not restrictive. Many variations of the invention may become apparent to those skilled in the art upon review of the disclosure. The scope of the invention can, therefore, be determined not with reference to the above description, but instead can be determined with reference to the pending claims along with their full scope or equivalents.

[0110] One or more features from any embodiment may be combined with one or more features of any other embodiment without departing from the scope of the invention.

[0111] A recitation of "a", "an" or "the" is intended to mean "one or more" unless specifically indicated to the contrary.

[0112] Although the invention has been described in detail for the purpose of illustration based on what is currently considered to be the most practical and preferred embodiments, it is to be understood that such detail is solely for that purpose and that the invention is not limited to the disclosed embodiments, but, on the contrary, is intended to cover modifications and equivalent arrangements that are within the spirit and scope of the appended claims. For example, it is to be understood that the present invention contemplates that, to the extent possible, one or more features of any embodiment can be combined with one or more features of any other embodiment.

[0113] While several inventive embodiments have been described and illustrated herein, those of ordinary skill in the art will readily envision a variety of other means and / or structures for performing the function and / or obtaining the results and / or one or more of the advantages described herein, and each of such variations and / or modifications is deemed to be within the scope of the inventive embodiments described herein. More generally, those skilled in the art will readily appreciate that all parameters, dimensions, materials, and configurations described herein are meant to be exemplary and that the actual parameters, dimensions, materials, and / or configurations will depend upon the specific application or applications for which the inventive teachings is / are used. Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific inventive embodiments described herein. It is, therefore, to be understood that the foregoing embodiments are presented by way of example only and that, within the scope of the appended claims and equivalents thereto, inventive embodiments may be practiced otherwise than as specifically described and claimed. Inventive embodiments of the present disclosure are directed to each individual feature, system, article, material, kit, and / or method described herein. In addition, any combination of two or more such features, systems, articles, materials, kits, and / or methods, if such features, systems, articles, materials, kits, and / or methods are not mutually inconsistent, is included within the inventive scope of the present disclosure.

Claims

We Claim:

1. A method (300) for predicting binding affinity of small molecules to Ribo-Nucleic Acid (RNA) targets, comprising: receiving (301), by a processor (201), an input from a user, comprising a sequence of an RNA target, information related to a small molecule and a sub-type of the RNA target; and predicting (302), by the processor (201), a binding affinity of the small molecule to the RNA target of the sub-type, using a prediction model (211), wherein generating the prediction model (211) comprises (303): obtaining (304), by the processor (201), a plurality of RNA sequence based features for a plurality of sub-types of a plurality of RNA targets, and a plurality of small molecule features for a plurality of small molecules; identifying (305), by the processor (201), a plurality of non-correlated features from the plurality of RNA sequence based features and the plurality of small molecule features; and generating (306), by the processor (201), a final feature combination from the plurality of non-correlated features for each sub-type of the plurality of sub-types of the plurality of RNA targets based on a quasi-binding affinity of each of the plurality of small molecules with each sub-type of the plurality of sub-types of the plurality of RNA targets, wherein the final feature combination is used to generate the prediction model (211), for respective sub-type.

2. The method (300) as claimed in claim 1, wherein the sub-types of the RNA target is one of aptamers, miRNA, repeats, ribosomal RNA, riboswitches or viral RNA.

3. The method (300) as claimed in claim 1, wherein the plurality of non-correlated features are identified using statistical methods.

4. The method (300) as claimed in claim 1, wherein generating (400) the prediction model (211) further comprises, iteratively performing, until identifying the final feature combination:inputting (401), by the processor (201), a feature combination of the plurality of small molecule features and the plurality of RNA sequence based features for each subtype, to the prediction model (211); determining, (402) by the processor (201), the quasi-binding affinity of the plurality of small molecules with each sub-type of the plurality of sub-types of the plurality of RNA targets; determining (403), by the processor (201), one or more performance variables for the quasi-binding affinity of the plurality of small molecules with each sub-type of the plurality of sub-types of the plurality of RNA targets; identifying (404), by the processor (201), the feature combination as the final feature combination when values of the one or more performance variables are within respective threshold range; incrementing (405), by the processor (201), number of features in the feature combination when the values of the one or more performance variables are outside respective threshold range; and generating (406), by the processor (201), the prediction model (211) to predict the binding affinity using the final feature combination.

5. The method (400) as claimed in claim 4, wherein the one or more performance variables comprises of Pearson Correlation Coefficient (r) and Mean Absolute Error (MAE).

6. A system (100) for predicting binding affinity of small molecules to Ribo-Nucleic Acid (RNA) targets, comprising of: a memory (202); and a processor (201), wherein the processor (201) is configured to: receive an input from a user, comprising a sequence of an RNA target, information related to a small molecule and a sub-type of the RNA target; and predict a binding affinity of the small molecule to the RNA target of the sub-type, using a prediction model (211), wherein the processor (201) configured to generate the prediction model (211), is configured to:obtain a plurality of RNA sequence based features for a plurality of sub-types of a plurality of RNA targets, and a plurality of small molecule features for a plurality of small molecules; identify a plurality of non-correlated features from the plurality of RNA sequence based features and the plurality of small molecule features; and generate a final feature combination from the plurality of non-correlated features for each sub-type of the plurality of sub-types of the plurality of RNA targets based on a quasi-binding affinity of each of the plurality of small molecules with each sub-type of the plurality of sub-types of the plurality of RNA targets, wherein the final feature combination is used to generate the prediction model (211).

7. The system (100) as claimed in claim 6, wherein the sub-types of the RNA target is one of aptamers, miRNA, repeats, ribosomal RNA, riboswitches or viral RNA.

8. The system (100) as claimed in claim 6, wherein the plurality of non-correlated features are identified using statistical methods.

9. The system (100) as claimed in claim 6, wherein the processor (201) configured to generate the prediction model (211) is further configured to, until the final feature combination is identified, iteratively: input a feature combination of the plurality of small molecule features and the plurality of RNA sequence based features for each sub-type, to the prediction model (211); determine the quasi-binding affinity of the plurality of small molecules with each sub-type of the plurality of sub-types of the plurality of RNA targets; determine one or more performance variables for the quasi-binding affinity of the plurality of small molecules with each sub-type of the plurality of sub-types of the plurality of RNA targets; identify the feature combination as the final feature combination when values of the one or more performance variables are within respective threshold range; increment number of features in the feature combination when the values of the one or more performance variables are outside respective threshold range; andgenerate the prediction model (211) to predict the binding affinity using the final feature combination.

10. The system (100) as claimed in claim 9, wherein the one or more performance variables comprises of Pearson Correlation Coefficient (r) and Mean Absolute Error (MAE).

Images