International Teaching | ARTIFICIAL INTELLIGENCE FOR OMICS DATA ANALYSIS

cod. 0623200011

ARTIFICIAL INTELLIGENCE FOR OMICS DATA ANALYSIS

	0623200011
	DEPARTMENT OF INFORMATION AND ELECTRICAL ENGINEERING AND APPLIED MATHEMATICS
	EQF7
	INFORMATION ENGINEERING FOR DIGITAL MEDICINE
	2024/2025

CURRICULUM ARTIFICIAL INTELLIGENCE FOR DIGITAL HEALTH

	OBBLIGATORIO
	YEAR OF COURSE 2
	YEAR OF DIDACTIC SYSTEM 2022
	AUTUMN SEMESTER

TEACHING UNITS

SSD	CFU	HOURS	ACTIVITY	TYPE OF ACTIVITY
ING-INF/05	3	24	LESSONS	COMPULSORY SUBJECTS, CHARACTERISTIC OF THE CLASS
ING-INF/05	2	16	EXERCISES	COMPULSORY SUBJECTS, CHARACTERISTIC OF THE CLASS
ING-INF/05	1	8	LAB	COMPULSORY SUBJECTS, CHARACTERISTIC OF THE CLASS

PROFESSORS

	DIEGO GRAGNANIELLO T Curriculum
	EDUARDO MARIA SOMMELLA Curriculum

	Objectives
	THE TEACHING DEEPENS THE METHODOLOGICAL AND TECHNOLOGICAL ASPECTS FOR THE DEFINITION OF ARTIFICIAL INTELLIGENCE MODELS AND ARCHITECTURE FOR PRECISION MEDICINE WITH EMPHASIS ON THE ANALYSIS OF OMIC DATASET FOR IDENTIFICATION, CLASSIFICATION AND STUDY OF PATHOLOGIES AS WELL AS THE EVALUATION OF DRUGS EFFECTIVENESS. KNOWLEDGE AND UNDERSTANDING SKILLS ARTIFICIAL INTELLIGENCE MODELS AND ARCHITECTURE FOR THE ANALYSIS, IDENTIFICATION AND CHARACTERIZATION OF PATHOLOGIES, OPERATING ON OMIC DATA OF VARIOUS KINDS AVAILABLE IN PUBLIC DATASETS TOGETHER WITH THE PRINCIPLES AND TECHNIQUES FOR A CORRECT CONFIGURATION, TUNING AND PERFORMANCE EVALUATION OF THE AI MODELS ADOPTED, KNOWLEDGE AND UNDERSTANDING SKILLS APPLIED BASED ON THE PROBLEM TO BE SOLVED, IDENTIFY THE MOST SUITABLE ARTIFICIAL INTELLIGENCE MODEL/ARCHITECTURE AND DESIGN, IMPLEMENT AND VALIDATE, SOFTWARE SOLUTIONS FOR PRECISION MEDICINE APPLYING THESE MODELS/ARCHITECTURE. DEFINE THE PARAMETERS FOR TRAINING, INTERPRETING AND EXPLAIN THE RESULTS PROVIDED BY THE MODELS.

THE TEACHING DEEPENS THE METHODOLOGICAL AND TECHNOLOGICAL ASPECTS FOR THE DEFINITION OF ARTIFICIAL INTELLIGENCE MODELS AND ARCHITECTURE FOR PRECISION MEDICINE WITH EMPHASIS ON THE ANALYSIS OF OMIC DATASET FOR IDENTIFICATION, CLASSIFICATION AND STUDY OF PATHOLOGIES AS WELL AS THE EVALUATION OF DRUGS EFFECTIVENESS.

KNOWLEDGE AND UNDERSTANDING SKILLS
ARTIFICIAL INTELLIGENCE MODELS AND ARCHITECTURE FOR THE ANALYSIS, IDENTIFICATION AND CHARACTERIZATION OF PATHOLOGIES, OPERATING ON OMIC DATA OF VARIOUS KINDS AVAILABLE IN PUBLIC DATASETS TOGETHER WITH THE PRINCIPLES AND TECHNIQUES FOR A CORRECT CONFIGURATION, TUNING AND PERFORMANCE EVALUATION OF THE AI MODELS ADOPTED,

KNOWLEDGE AND UNDERSTANDING SKILLS APPLIED
BASED ON THE PROBLEM TO BE SOLVED, IDENTIFY THE MOST SUITABLE ARTIFICIAL INTELLIGENCE MODEL/ARCHITECTURE AND DESIGN, IMPLEMENT AND VALIDATE, SOFTWARE SOLUTIONS FOR PRECISION MEDICINE APPLYING THESE MODELS/ARCHITECTURE. DEFINE THE PARAMETERS FOR TRAINING, INTERPRETING AND EXPLAIN THE RESULTS PROVIDED BY THE MODELS.

	Prerequisites
	COMPUTATIONAL GENOMICS AND MACHINE LEARNING COURSES ARE PREREQUISITE FOR THIS COURSE.

	Contents
	DIDACTIC UNIT 1: STATISTICAL MACHINE LEARNING MODELS (LECTURE/PRACTICE/LABORATORY HOURS 8/4/2) - 1 (2 HOURS LECTURE): INTRODUCTION TO LINEAR AND SVM CLASSIFIERS - 2 (2 HOURS LECTURE): LINEAR REGRESSION AND LOGISTICS REGRESSION - 3 (2 HOURS LECTURE): NOTES WITH MULTICLASSIFIER SYSTEMS BASED ON DECISION TREE - 4 (2 HOURS LECTURE): NOTES WITH GRAPH NEURAL NETWORKS - 4 (4 HOURS PRACTICE): PYTHON APPLICATION EXAMPLES: SCIKIT-LEARN AND PYTORCH - 6 (2 HOURS LABORATORY): APPLICATION TO REAL-WORLD PROBLEMS KNOWLEDGE AND ABILITY TO UNDERSTAND: CHARACTERISTICS OF THE MODELS, CONFIGURATION AND TUNING KNOWLEDGE AND UNDERSTANDING SKILLS APPLIED: USE THE RIGHT MODEL BASED ON THE PROBLEM TO BE SOLVED, CONFIGURE IT CORRECTLY AND INTERPRET THE ACHIEVED RESULTS DIDACTIC UNIT 2: WORKING WITH OMIC DATA (LECTURE/PRACTICE/LABORATORY HOURS 12/10/0) - 7 (4/2 HOURS LECTURE/PRACTICE): GENOMICS/EPIGENOMICS: DATASES, THEIR USE FOR DIAGNOSIS AND CARE, APPLICATION EXAMPLES - 8 (4/2 HOURS LECTURE/PRACTICE): PROTEOMICS: DATASES, THEIR USE FOR DIAGNOSIS AND8 CARE, APPLICATION EXAMPLES - 9 (4/2 HOURS LECTURE/PRACTICE): METABOLOMICS: DATASES, THEIR USE FOR DIAGNOSIS AND CARE, APPLICATION EXAMPLES - 10 (4 HOURS PRACTICE): EXAMPLES OF APPLICATION OF ML/DL KNOWLEDGE AND ABILITY TO UNDERSTAND: WHAT TYPE OF DATASET TO USE ACCORDING TO THE ANALYSIS TO BE CARRIED OUT KNOWLEDGE AND UNDERSTANDING SKILLS APPLIED: USE THE RESOURCES AVAILABLE ON THE NET AND APPLY ML MODELS FOR DATA ANALYSIS AND EXPLAIN/MOTIVATE THE RESULTS DIDACTIC UNIT 3: ML AND DL WITH N << P (LECTURE/PRACTICE HOURS/LABORATORY 4/2/0) - 11 (2 HOURS LECTURE): CLUSTERING TECHNIQUES AND UNSUPERVISED LEARNING - 12 (2 HOURS LECTURE): PCA, K-MEANS, T-SNE - 13 (2 HOURS LECTURE): ANOMALY DETECTION - 14 (2 HOURS PRACTICE): OVERSAMPLING AND UNDERSAMPLING TECHNIQUES KNOWLEDGE AND ABILITY TO UNDERSTAND: TECHNIQUES FOR MANAGING N << P AND INTERPRETATION/EXPLANATION OF THE RESULTS KNOWLEDGE AND UNDERSTANDING SKILLS APPLIED: USE OF THE MOST APPROPRIATE TECHNIQUES FOR N << P (SELECTION/REDUCTION UNSUPERVISED; DATA AUGMENTATION, OUTLIERS REMOVAL, ETC.) DIDACTIC UNIT 4: PROJECT WORK (LECTURE/PRACTICE/LABORATORY HOURS 0/0/4) KNOWLEDGE AND UNDERSTANDING SKILLS: ANALYSIS OF THE PROBLEM TO BE ADDRESSED AND IDENTIFIED OF THE DATA, MODELS AND TECHNIQUES MOST APPROPRIATE KNOWLEDGE AND UNDERSTANDING SKILLS APPLIED: IMPLEMENTATION , VALIDATION AND INTERPRETATION OF THE RESULTS ACHIEVED TOTAL HOURS LECTURE/PRACTICE/LABORATORY 26/16/6

Contents

DIDACTIC UNIT 1: STATISTICAL MACHINE LEARNING MODELS (LECTURE/PRACTICE/LABORATORY HOURS 8/4/2)
- 1 (2 HOURS LECTURE): INTRODUCTION TO LINEAR AND SVM CLASSIFIERS
- 2 (2 HOURS LECTURE): LINEAR REGRESSION AND LOGISTICS REGRESSION
- 3 (2 HOURS LECTURE): NOTES WITH MULTICLASSIFIER SYSTEMS BASED ON DECISION TREE
- 4 (2 HOURS LECTURE): NOTES WITH GRAPH NEURAL NETWORKS
- 4 (4 HOURS PRACTICE): PYTHON APPLICATION EXAMPLES: SCIKIT-LEARN AND PYTORCH
- 6 (2 HOURS LABORATORY): APPLICATION TO REAL-WORLD PROBLEMS

KNOWLEDGE AND ABILITY TO UNDERSTAND: CHARACTERISTICS OF THE MODELS, CONFIGURATION AND TUNING
KNOWLEDGE AND UNDERSTANDING SKILLS APPLIED: USE THE RIGHT MODEL BASED ON THE PROBLEM TO BE SOLVED, CONFIGURE IT CORRECTLY AND INTERPRET THE ACHIEVED RESULTS

DIDACTIC UNIT 2: WORKING WITH OMIC DATA (LECTURE/PRACTICE/LABORATORY HOURS 12/10/0)
- 7 (4/2 HOURS LECTURE/PRACTICE): GENOMICS/EPIGENOMICS: DATASES, THEIR USE FOR DIAGNOSIS AND CARE, APPLICATION EXAMPLES
- 8 (4/2 HOURS LECTURE/PRACTICE): PROTEOMICS: DATASES, THEIR USE FOR DIAGNOSIS AND8 CARE, APPLICATION EXAMPLES
- 9 (4/2 HOURS LECTURE/PRACTICE): METABOLOMICS: DATASES, THEIR USE FOR DIAGNOSIS AND CARE, APPLICATION EXAMPLES
- 10 (4 HOURS PRACTICE): EXAMPLES OF APPLICATION OF ML/DL

KNOWLEDGE AND ABILITY TO UNDERSTAND: WHAT TYPE OF DATASET TO USE ACCORDING TO THE ANALYSIS TO BE CARRIED OUT
KNOWLEDGE AND UNDERSTANDING SKILLS APPLIED: USE THE RESOURCES AVAILABLE ON THE NET AND APPLY ML MODELS FOR DATA ANALYSIS AND EXPLAIN/MOTIVATE THE RESULTS

DIDACTIC UNIT 3: ML AND DL WITH N << P (LECTURE/PRACTICE HOURS/LABORATORY 4/2/0)
- 11 (2 HOURS LECTURE): CLUSTERING TECHNIQUES AND UNSUPERVISED LEARNING
- 12 (2 HOURS LECTURE): PCA, K-MEANS, T-SNE
- 13 (2 HOURS LECTURE): ANOMALY DETECTION
- 14 (2 HOURS PRACTICE): OVERSAMPLING AND UNDERSAMPLING TECHNIQUES

KNOWLEDGE AND ABILITY TO UNDERSTAND: TECHNIQUES FOR MANAGING N << P AND INTERPRETATION/EXPLANATION OF THE RESULTS
KNOWLEDGE AND UNDERSTANDING SKILLS APPLIED: USE OF THE MOST APPROPRIATE TECHNIQUES FOR N << P (SELECTION/REDUCTION UNSUPERVISED; DATA AUGMENTATION, OUTLIERS REMOVAL, ETC.)

DIDACTIC UNIT 4: PROJECT WORK (LECTURE/PRACTICE/LABORATORY HOURS 0/0/4)
KNOWLEDGE AND UNDERSTANDING SKILLS: ANALYSIS OF THE PROBLEM TO BE ADDRESSED AND IDENTIFIED OF THE DATA, MODELS AND TECHNIQUES MOST APPROPRIATE
KNOWLEDGE AND UNDERSTANDING SKILLS APPLIED: IMPLEMENTATION , VALIDATION AND INTERPRETATION OF THE RESULTS ACHIEVED

TOTAL HOURS LECTURE/PRACTICE/LABORATORY 26/16/6

	Teaching Methods
	THE TEACHING ACTIVITIES (48H OF LESSONS, EXERCISES AND LABORATORY ACTIVITIES) ARE CHARACTERIZED BY A DYNAMIC SETTING, WHICH INCLUDES THE ANALYSIS OF CASE STUDIES WITH THE ACTIVE PARTICIPATION OF STUDENTS WHO, DURING PRACTICAL AND LABORATORY ACTIVITIES, WILL CARRY OUT SPECIFIC INSIGHTS ON THE FEATURES OF MACHINE LEARNING MODELS AND THEIR CONFIGURATION IN ORDER TO BE USED FOR OMIC DATA ANALYSIS. IN PARTICULAR, THE EDUCATIONAL ACTIVITIES WILL INCLUDE LESSONS (18 HOURS), PRACTICES (20 HOURS) AND LABORATORY (10 HOURS) WHICH ALSO INCLUDES THE DEVELOPMENT OF THE PROJECT WORK. FOR THE DEVELOPMENT OF THE PROJECT STUDENTS WILL APPLY THE KNOWLEDGE ACQUIRED IN ORDER TO SELF-CHOOSE THE MOST APPROPRIATE MODELS AND FRAMEWORKS TO SOLVE SPECIFIC PROBLEMS IN THE APPLICATION DOMAINS PRESENTED DURING THE COURSE. EDUCATIONAL ACTIVITIES WILL BE SUPPORTED BY THE USE OF THE UNIVERSITY E-LEARNING PLATFORM TO FACILITATE AND STIMULATE DISCUSSION AND DEBATE AMONG STUDENTS, AS WELL AS FOR NOTIFICATION AND DISTRIBUTION OF EDUCATIONAL MATERIAL.

Teaching Methods

THE TEACHING ACTIVITIES (48H OF LESSONS, EXERCISES AND LABORATORY ACTIVITIES) ARE CHARACTERIZED BY A DYNAMIC SETTING, WHICH INCLUDES THE ANALYSIS OF CASE STUDIES WITH THE ACTIVE PARTICIPATION OF STUDENTS WHO, DURING PRACTICAL AND LABORATORY ACTIVITIES, WILL CARRY OUT SPECIFIC INSIGHTS ON THE FEATURES OF MACHINE LEARNING MODELS AND THEIR CONFIGURATION IN ORDER TO BE USED FOR OMIC DATA ANALYSIS.
IN PARTICULAR, THE EDUCATIONAL ACTIVITIES WILL INCLUDE LESSONS (18 HOURS), PRACTICES (20 HOURS) AND LABORATORY (10 HOURS) WHICH ALSO INCLUDES THE DEVELOPMENT OF THE PROJECT WORK.
FOR THE DEVELOPMENT OF THE PROJECT STUDENTS WILL APPLY THE KNOWLEDGE ACQUIRED IN ORDER TO SELF-CHOOSE THE MOST APPROPRIATE MODELS AND FRAMEWORKS TO SOLVE SPECIFIC PROBLEMS IN THE APPLICATION DOMAINS PRESENTED DURING THE COURSE.
EDUCATIONAL ACTIVITIES WILL BE SUPPORTED BY THE USE OF THE UNIVERSITY E-LEARNING PLATFORM TO FACILITATE AND STIMULATE DISCUSSION AND DEBATE AMONG STUDENTS, AS WELL AS FOR NOTIFICATION AND DISTRIBUTION OF EDUCATIONAL MATERIAL.

	Verification of learning
	THE FINAL EXAM IS DESIGNED TO ASSESS THE OVERALL KNOWLEDGE AND UNDERSTANDING OF THE CONCEPTS PRESENTED IN THE COURSE, THE ABILITY TO APPLY THAT KNOWLEDGE TO DEVELOP SPECIFIC APPLICATIONS AS WELL AS THE ABILITY TO COMMUNICATE AND PRESENT THE WORK CARRIED OUT (COMMUNICATION SKILLS). THE EXAMINATION CONSISTS OF A PRACTICAL PART AND AN ORAL EXAM (INTERVIEW). THE PRACTICAL PART CONSISTS OF THE DEVELOPMENT OF A PROJECT WORK TO BE CARRIED OUT IN GROUPS (2-4 STUDENTS) ON THE PROPOSED TOPICS. THE ORAL EXAM CONSISTS OF THE PRESENTATION OF WHAT HAS BEEN ACHIEVED DURING THE DEVELOPMENT OF THE PROJECT WORK. EACH GROUP MEMBERS EXPOSE ITS OWN CONTRIBUTION FOR THE REALIZATION OF THE PROJECT TOGETHER WITH A DISCUSSION OF THE TOOLS AND FRAMEWORK USED, THE SOFTWARE ARCHITECTURE AND THE ACHIEVED RESULTS. DURING THE ORAL THE ORAL EXAM WILL BE ALSO USED TO ASSEST THE STUDENT KNOWLEDGE ABOUT THE TOPICS PRESENTED DURING THE COURSE. IN THE FINAL EVALUATION, EXPRESSED WITH A MARK RANGE OF 30/30, THE PRACTICAL PART WILL WEIGH 65% AND THE ORAL EXAM FOR 35%. “HONOURS” (30/30 CUM LAUDE) WILL BE AWARDED TO STUDENTS WHO DEMONSTRATE A FULL MASTERY OF ALL THE MAIN METHODOLOGICAL AND TECHNOLOGICAL ASPECTS ADDRESSED IN THE COURSE AND HOW THEY CAN BE USED FOR THE CREATION OF APPLICATIONS AND SOLUTIONS IN DIFFERENT APPLICATION DOMAINS TOGETHER WITH THE IMPLICATIONS DERIVED FROM THEIR USE.

Verification of learning

THE FINAL EXAM IS DESIGNED TO ASSESS THE OVERALL KNOWLEDGE AND UNDERSTANDING OF THE CONCEPTS PRESENTED IN THE COURSE, THE ABILITY TO APPLY THAT KNOWLEDGE TO DEVELOP SPECIFIC APPLICATIONS AS WELL AS THE ABILITY TO COMMUNICATE AND PRESENT THE WORK CARRIED OUT (COMMUNICATION SKILLS).
THE EXAMINATION CONSISTS OF A PRACTICAL PART AND AN ORAL EXAM (INTERVIEW). THE PRACTICAL PART CONSISTS OF THE DEVELOPMENT OF A PROJECT WORK TO BE CARRIED OUT IN GROUPS (2-4 STUDENTS) ON THE PROPOSED TOPICS.
THE ORAL EXAM CONSISTS OF THE PRESENTATION OF WHAT HAS BEEN ACHIEVED DURING THE DEVELOPMENT OF THE PROJECT WORK.
EACH GROUP MEMBERS EXPOSE ITS OWN CONTRIBUTION FOR THE REALIZATION OF THE PROJECT TOGETHER WITH A DISCUSSION OF THE TOOLS AND FRAMEWORK USED, THE SOFTWARE ARCHITECTURE AND THE ACHIEVED RESULTS. DURING THE ORAL THE ORAL EXAM WILL BE ALSO USED TO ASSEST THE STUDENT KNOWLEDGE ABOUT THE TOPICS PRESENTED DURING THE COURSE.
IN THE FINAL EVALUATION, EXPRESSED WITH A MARK RANGE OF 30/30, THE PRACTICAL PART WILL WEIGH 65% AND THE ORAL EXAM FOR 35%. “HONOURS” (30/30 CUM LAUDE) WILL BE AWARDED TO STUDENTS WHO DEMONSTRATE A FULL MASTERY OF ALL THE MAIN METHODOLOGICAL AND TECHNOLOGICAL ASPECTS ADDRESSED IN THE COURSE AND HOW THEY CAN BE USED FOR THE CREATION OF APPLICATIONS AND SOLUTIONS IN DIFFERENT APPLICATION DOMAINS TOGETHER WITH THE IMPLICATIONS DERIVED FROM THEIR USE.

	Texts
	HANDOUTS AND SUPPLEMENTARY MATERIAL PROVIDED BY THE TEACHER WILL BE AVAILABLE ON THE UNIVERSITY E-LEARNING PLATFORM (HTTP://ELEARNING.UNISA.IT) ACCESSIBLE TO STUDENTS USING THEIR OWN UNIVERSITY CREDENTIALS. ARE ALSO SUGGESTED THE FOLLOWING BOOKS AURÉLIEN GÉRON, "HANDS-ON MACHINE LEARNING WITH SCIKIT-LEARN KERAS AND TENSORFLOW“, O REILLY ED. GARETH JAMES, DANIELA WITTEN, TREVOR HASTIE, ROBERT TIBSHIRANI AN INTRODUCTION TO STATISTICAL LEARNING, SPRINGER

Texts

HANDOUTS AND SUPPLEMENTARY MATERIAL PROVIDED BY THE TEACHER WILL BE AVAILABLE ON THE UNIVERSITY E-LEARNING PLATFORM (HTTP://ELEARNING.UNISA.IT) ACCESSIBLE TO STUDENTS USING THEIR OWN UNIVERSITY CREDENTIALS.

ARE ALSO SUGGESTED THE FOLLOWING BOOKS

AURÉLIEN GÉRON, "HANDS-ON MACHINE LEARNING WITH SCIKIT-LEARN KERAS AND TENSORFLOW“, O REILLY ED.
GARETH JAMES, DANIELA WITTEN, TREVOR HASTIE, ROBERT TIBSHIRANI
AN INTRODUCTION TO STATISTICAL LEARNING, SPRINGER

	More Information
	THE COURSE IS HELD IN ENGLISH

Lessons Timetable

BETA VERSION Data source ESSE3

International Teaching | ARTIFICIAL INTELLIGENCE FOR OMICS DATA ANALYSIS