MultiSec: Multi-Task Deep Learning Improves Secreted Protein Discovery in Human Body Fluids
Prediction of secreted proteins in human body fluids is essential since secreted proteins hold promise as disease biomarkers. Various approaches have been proposed to predict whether a protein is secreted into a specific fluid by its sequence. However, there may be relationships between different hu...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-07-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/10/15/2562 |
| _version_ | 1797441491008749568 |
|---|---|
| author | Kai He Yan Wang Xuping Xie Dan Shao |
| author_facet | Kai He Yan Wang Xuping Xie Dan Shao |
| author_sort | Kai He |
| collection | DOAJ |
| description | Prediction of secreted proteins in human body fluids is essential since secreted proteins hold promise as disease biomarkers. Various approaches have been proposed to predict whether a protein is secreted into a specific fluid by its sequence. However, there may be relationships between different human body fluids when proteins are secreted into these fluids. Current approaches ignore these relationships directly, and therefore their performances are limited. Here, we present MultiSec, an improved approach for secreted protein discovery to exploit relationships between fluids via multi-task learning. Specifically, a sampling-based balance strategy is proposed to solve imbalance problems in all fluids, an effective network is presented to extract features for all fluids, and multi-objective gradient descent is employed to prevent fluids from hurting each other. MultiSec was trained and tested in 17 human body fluids. The comparison benchmarks on the independent testing datasets demonstrate that our approach outperforms other available approaches in all compared fluids. |
| first_indexed | 2024-03-09T12:23:51Z |
| format | Article |
| id | doaj.art-8719fe177df744a582d6fe1acf62f9cc |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-09T12:23:51Z |
| publishDate | 2022-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-8719fe177df744a582d6fe1acf62f9cc2023-11-30T22:37:15ZengMDPI AGMathematics2227-73902022-07-011015256210.3390/math10152562MultiSec: Multi-Task Deep Learning Improves Secreted Protein Discovery in Human Body FluidsKai He0Yan Wang1Xuping Xie2Dan Shao3Key Laboratory of Symbol Computation and Knowledge Engineering of Ministry of Education, College of Computer Science and Technology, Jilin University, Changchun 130012, ChinaKey Laboratory of Symbol Computation and Knowledge Engineering of Ministry of Education, College of Computer Science and Technology, Jilin University, Changchun 130012, ChinaKey Laboratory of Symbol Computation and Knowledge Engineering of Ministry of Education, College of Computer Science and Technology, Jilin University, Changchun 130012, ChinaCollege of Computer Science and Technology, Changchun University, Changchun 130022, ChinaPrediction of secreted proteins in human body fluids is essential since secreted proteins hold promise as disease biomarkers. Various approaches have been proposed to predict whether a protein is secreted into a specific fluid by its sequence. However, there may be relationships between different human body fluids when proteins are secreted into these fluids. Current approaches ignore these relationships directly, and therefore their performances are limited. Here, we present MultiSec, an improved approach for secreted protein discovery to exploit relationships between fluids via multi-task learning. Specifically, a sampling-based balance strategy is proposed to solve imbalance problems in all fluids, an effective network is presented to extract features for all fluids, and multi-objective gradient descent is employed to prevent fluids from hurting each other. MultiSec was trained and tested in 17 human body fluids. The comparison benchmarks on the independent testing datasets demonstrate that our approach outperforms other available approaches in all compared fluids.https://www.mdpi.com/2227-7390/10/15/2562secreted protein discoverymulti-task learningdeep learning |
| spellingShingle | Kai He Yan Wang Xuping Xie Dan Shao MultiSec: Multi-Task Deep Learning Improves Secreted Protein Discovery in Human Body Fluids Mathematics secreted protein discovery multi-task learning deep learning |
| title | MultiSec: Multi-Task Deep Learning Improves Secreted Protein Discovery in Human Body Fluids |
| title_full | MultiSec: Multi-Task Deep Learning Improves Secreted Protein Discovery in Human Body Fluids |
| title_fullStr | MultiSec: Multi-Task Deep Learning Improves Secreted Protein Discovery in Human Body Fluids |
| title_full_unstemmed | MultiSec: Multi-Task Deep Learning Improves Secreted Protein Discovery in Human Body Fluids |
| title_short | MultiSec: Multi-Task Deep Learning Improves Secreted Protein Discovery in Human Body Fluids |
| title_sort | multisec multi task deep learning improves secreted protein discovery in human body fluids |
| topic | secreted protein discovery multi-task learning deep learning |
| url | https://www.mdpi.com/2227-7390/10/15/2562 |
| work_keys_str_mv | AT kaihe multisecmultitaskdeeplearningimprovessecretedproteindiscoveryinhumanbodyfluids AT yanwang multisecmultitaskdeeplearningimprovessecretedproteindiscoveryinhumanbodyfluids AT xupingxie multisecmultitaskdeeplearningimprovessecretedproteindiscoveryinhumanbodyfluids AT danshao multisecmultitaskdeeplearningimprovessecretedproteindiscoveryinhumanbodyfluids |