This dataset for musculoskeletal pain research contains 6 h and 4 min of biosignals acquired from 17 healthy participants with delayed-onset muscle soreness (DOMS) and 1 h 6 min from 6 participants with shoulder musculoskeletal disorders (MSDs), performing industrial tasks, where certain movements trigger musculoskeletal pain in the shoulder/upper arm. In each trial, several biosignals were recorded: Electrocardiogram (ECG), sampled at 1259.26 Hz, using the Trigno EKG Biofeedback Sensor (Delsys Incorporated); Surface electromyography (sEMG) from the sore upper-arm muscles biceps brachii, deltoid anterior, deltoid medius, and deltoid posterior, sampled at 2148.15 Hz, using the Trigno Avanti (Delsys Incorporated); Inertial data from the upper limbs, sampled at 60 Hz, using the Xsens MTw Awinda + Xsens MVN Analyze (Movella Inc.); Participants’ self-reported pain level, which was defined as binary (0 - no pain, 1 - pain; 2 - samples to discard), sampled at 100 Hz. Database structure: Protocol: step-by-step description of the acquisition protocol. Code: Arduino file to acquire the labels (labelling_buttons.ino), Jupyter Notebook files containing the base code to read and compute physiological features (feature_extraction_ECG.ipynb, feature_extraction_EMG.ipynb, feature_extraction_IMU.ipynb), and Jupyter Notebook file to perform a statistical analysis (analysis_features.ipynb). DOMS dataset.zip: includes data (biosignals and pain labels acquired in each trial for each participant, stored in .csv and .xlsx files) and metadata (participants' anthropometric data, cardiac conditions, anti-inflammatories, caffeine, alcohol and nicotine intake, and exercise habits); MSD dataset.zip: includes data (biosignals, pain labels and task labels acquired in each trial for each participant, stored in .csv and .xlsx files) and metadata (participants' anthropometric data, cardiac conditions, anti-inflammatories, caffeine, alcohol and nicotine intake, exercise habits, musculoskeletal disorder description, and physiotherapy info). This dataset may contribute to the development and testing of new pain detection algorithms and analysis of the underlying mechanisms. For any questions, please contact Diogo R. Martins at diogo-martins-9@live.com.pt, Sara M. Cerqueira at saracerqueira1996@gmail.com or Cristina P. Santos at cristina@dei.uminho.pt.