Documentation

Comprehensive guides and documentation for all Geinforce Technology tools and services.

TABLE OF CONTENTS

Getting Started

Welcome to Geinforce Technology's documentation. This guide will help you get started with our suite of computational tools for drug discovery.

Account Setup

Create an account and get started with Geinforce Technology's platform.

  1. Visit our registration page
  2. Enter your details and verify your email
  3. Explore available tools and plans
  4. Start using our computational tools
Create Account
Quick Start Guide

Jump right in with our most popular tools:

View Tutorials

Tools Overview

Geinforce Technology offers a comprehensive suite of computational tools designed to accelerate drug discovery research. Each tool is built on cutting-edge algorithms and trained on extensive datasets to provide accurate and reliable results.

ForceADME

Predict ADME properties (Absorption, Distribution, Metabolism, Excretion) for small molecules to evaluate their drug-likeness.

GeinDock Suite

Advanced molecular docking platform for predicting binding affinities and interactions between ligands and protein targets.

ForcePhrase

Natural language processing tool for extracting insights from scientific literature and patents related to drug discovery.

AI Detection

Analyze text to determine if it was generated by AI, helping maintain integrity in scientific publications.

ForceADME

ForceADME is our advanced ADME property prediction tool that helps researchers evaluate the drug-likeness of small molecules. It provides comprehensive analysis of Absorption, Distribution, Metabolism, and Excretion properties.

Key Features

  • Comprehensive Property Prediction: Calculate over 30 different molecular and ADME properties
  • Multiple LogP Calculation Methods: Including consensus LogP, XLogP3, WLogP, MLogP, and more
  • Water Solubility Estimation: Using ESOL, Ali, and SILICOS-IT methods
  • Pharmacokinetics Predictions: GI absorption, BBB permeation, P-gp substrate, CYP inhibition
  • Drug-likeness Analysis: Lipinski, Ghose, Veber, Egan, and Muegge rule evaluations
  • Medicinal Chemistry Alerts: PAINS and Brenk structural alerts
  • Interactive Visualization: Including Boiled Egg plot for absorption and BBB permeation

Technology

ForceADME uses a combination of physics-based methods and machine learning approaches:

  • RDKit-based molecular property calculations
  • Consensus approaches for improved accuracy
  • Deep neural networks for complex property predictions
  • QSPR (Quantitative Structure-Property Relationship) models

Getting Started with ForceADME

Follow these steps to analyze a molecule using ForceADME:

  1. Access ForceADME: Navigate to the ForceADME tool in your dashboard.

  2. Input Your Molecule: You can input your molecule in several ways:

    • Draw the molecule using the built-in JSME editor
    • Enter the SMILES string directly
    • Upload a file containing the structure (SDF, MOL)

  3. Run Analysis: Click the "Analyze" button to start the computation.

  4. View Results: Results will be displayed in an interactive dashboard with detailed sections for each property type.

  5. Export Results: You can export the results in various formats including PDF, Excel, and JSON.

Tip: For batch processing of multiple compounds, use our API or the batch upload feature in the web interface.

Example Use Cases

Example 1: Analyzing Aspirin

SMILES: CC(=O)OC1=CC=CC=C1C(=O)O

Key predicted properties:

  • Molecular Weight: 180.16 g/mol
  • LogP: 1.31
  • Topological Polar Surface Area: 63.60 Ų
  • H-Bond Donors: 1
  • H-Bond Acceptors: 4
  • GI Absorption: High
  • BBB Permeant: Yes
  • Bioavailability Score: 0.85

Interpretation: Aspirin passes all drug-likeness rules and has favorable ADME properties, which aligns with its well-known pharmacokinetic profile.

Example 2: Comparing Drug Candidates

ForceADME can be used to compare multiple drug candidates to identify the most promising lead compounds:

  1. Analyze each candidate molecule
  2. Compare key ADME parameters
  3. Identify compounds with the best overall profile
  4. Export comparison reports for team discussion

This approach helps prioritize the most promising candidates early in the development process, saving time and resources.

Input Parameters

Parameter Description Required Default
SMILES SMILES string representation of the molecule Yes None
Molecule Name Custom name for the molecule No "Unnamed Compound"
Calculation Mode Standard or Advanced mode for property calculation No Standard

Output Parameters

ForceADME provides a comprehensive set of calculated properties, including:

Molecular Properties
  • Molecular Formula
  • Molecular Weight
  • Heavy Atom Count
  • Aromatic Heavy Atoms
  • Fraction Csp3
  • Rotatable Bonds
  • H-Bond Acceptors
  • H-Bond Donors
  • Molar Refractivity
  • TPSA
Pharmacokinetics
  • GI Absorption
  • BBB Permeant
  • P-gp Substrate
  • CYP1A2/2C19/2C9/2D6/3A4 Inhibition
  • Log Kp (skin permeation)
Drug Likeness
  • Lipinski Rule of Five
  • Ghose Filter
  • Veber Rules
  • Egan (BBB) Rule
  • Bioavailability Score

GeinDock Suite

GeinDock Suite provides advanced molecular docking capabilities for predicting binding affinities and interactions between small molecules and protein targets.

Features Overview

  • Flexible docking engine with multiple scoring functions
  • Support for diverse protein targets and ligand types
  • Automated binding site detection
  • Batch processing capabilities
  • Interactive 3D visualization of docking results
  • Integrated with protein preparation tools
  • Customizable docking parameters
  • Comprehensive binding interaction analysis
  • Export results in various formats
  • Access to curated protein structure database

How It Works

GeinDock Workflow
Detailed documentation for GeinDock Suite is available.
For complete usage instructions, examples, and API documentation, please refer to the GeinDock User Manual.

ForcePhrase

ForcePhrase is our natural language processing tool designed specifically for extracting insights from scientific literature and patents related to drug discovery.

Key Capabilities

Smart Literature Search

Semantic search capabilities across scientific databases, patents, and publications.

Relationship Extraction

Identify compound-target, drug-disease, and other biomedical relationships.

Knowledge Graphs

Build and visualize complex biomedical knowledge networks from literature.

Under Development: ForcePhrase is currently in beta. Full documentation will be available when the tool is officially released.

AI Detection

Our AI Detection tool helps identify content that was likely generated by artificial intelligence systems, helping maintain integrity in scientific publications and other written content.

Key Features

  • High accuracy detection of AI-generated text
  • Support for multiple languages
  • Analyzes writing style, patterns, and linguistic markers
  • Provides confidence scores and detailed analysis
  • Continuously updated to detect output from latest AI models

How to Use

  1. Navigate to the AI Detection tool
  2. Input the text you want to analyze (paste directly or upload a file)
  3. Click "Analyze" to process the text
  4. Review the results, including the AI probability score and analysis

Example Analysis

Here's an example of how our AI Detection tool analyzes text:

Input Text: "The binding affinity of the compound to the target protein was determined to be in the nanomolar range, suggesting a high potential for therapeutic application. Further studies are needed to evaluate its pharmacokinetic properties and in vivo efficacy."
Analysis Results
15%

Verdict: Likely human-written

Confidence Factors
  • Specific scientific terminology
  • Natural sentence variation
  • Domain-appropriate phrasing
  • Presence of nuanced qualifiers

GeinBioPredictor

GeinBiopredictor is our tool for bioactivity prediction, designed to estimate the activity of compounds against specific targets using advanced machine learning algorithms.

Features

  • Predict bioactivity against hundreds of protein targets
  • Estimate IC50, EC50, Ki, and Kd values
  • Identify potential off-target effects
  • Analyze structure-activity relationships
  • Generate bioactivity profiles for lead optimization
Coming Soon: Detailed documentation for GeinBiopredictor will be available in the next release.

API Reference

Geinforce Technology provides comprehensive APIs for integrating our tools into your workflows and applications. This section covers the basics of our API usage.

Authentication

All API requests must be authenticated using API keys. To get your API key:

  1. Log in to your Geinforce account
  2. Navigate to Account Settings > API Keys
  3. Generate a new API key
curl -X GET "https://api.geinforce.com/v1/adme/analyze" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json"
Security Note: Keep your API key secure and never share it in client-side code.

ADME API

The ADME API allows you to programmatically analyze compounds for their ADME properties.

POST /v1/adme/analyze

Analyze a compound for ADME properties.

Request Parameters

smiles (string, required): SMILES string of the molecule

name (string, optional): Custom name for the molecule

mode (string, optional): Calculation mode ("standard" or "advanced")

Example Request
{
  "smiles": "CC(=O)OC1=CC=CC=C1C(=O)O",
  "name": "Aspirin",
  "mode": "advanced"
}
Example Response
{
  "id": "a1b2c3d4-e5f6-7g8h-9i0j-klmnopqrstuv",
  "name": "Aspirin",
  "smiles": "CC(=O)OC1=CC=CC=C1C(=O)O",
  "properties": {
    "molecular_formula": "C9H8O4",
    "molecular_weight": 180.16,
    "logp": 1.31,
    // Additional properties...
  }
}

Docking API

The Docking API provides programmatic access to our molecular docking engine.

Complete API Documentation: For the full API documentation including all endpoints, parameters, and response formats, please refer to our API Documentation Portal.

SDKs & Libraries

Geinforce provides software development kits (SDKs) in several popular programming languages to make integration with our APIs easier.

Python SDK

Our Python SDK is ideal for data scientists and computational chemists working in Python environments.

pip install geinforce-sdk
Documentation GitHub

R Package

Our R package provides easy access to Geinforce APIs from R statistical environment.

install.packages("geinforceR")
Documentation GitHub

Frequently Asked Questions

Our platform supports various molecular formats including SMILES, SMARTS, InChI, MOL, SDF, PDB, and more. The web interface primarily uses SMILES and file uploads (MOL, SDF), while the API has broader format support.

ForceADME's prediction accuracy varies by property. Basic physicochemical properties (LogP, MW, TPSA) are highly accurate as they are calculated from first principles. For more complex properties like bioavailability and GI absorption, our models achieve 80-85% accuracy in validation studies using external datasets. We provide confidence scores with predictions where applicable.

Yes, API usage limits depend on your subscription plan. Free accounts have a limit of 100 requests per day, while paid plans offer higher limits (Basic: 1,000/day, Professional: 10,000/day, Enterprise: custom limits). You can view your current usage and limits in your account dashboard.

Yes, GeinDock Suite allows you to upload your own protein structures in PDB format. You can also use structures from our curated database of over 10,000 prepared protein targets. Custom proteins go through an automated preparation pipeline to ensure they are ready for docking.

We take data security very seriously. All data is encrypted both in transit (using TLS) and at rest. We do not use customer data to train our models without explicit permission. Enterprise customers can opt for private cloud deployments with additional security measures. For more details, please refer to our Privacy Policy.

Troubleshooting

Encountering issues? Here are solutions to common problems.

Common Issues & Solutions
Invalid SMILES Error

Problem: Receiving an "Invalid SMILES string" error when analyzing a molecule.

Solution: Check for syntax errors in your SMILES string. Make sure valences are correct and all atoms are properly defined. You can validate your SMILES using our SMILES validator tool or third-party services.

API Authentication Failed

Problem: Receiving a 401 Unauthorized error when making API calls.

Solution: Verify that you're using the correct API key and that it's active. Make sure you're including the Authorization header correctly. Check if your subscription plan is active and has sufficient API credits remaining.

Docking Job Fails

Problem: GeinDock job fails with an error.

Solution: Check that your protein structure is complete and properly prepared. If using a custom ligand, verify its structure. For large jobs, try splitting into smaller batches. If issues persist, contact support with your job ID for assistance.

Need Additional Help?

If you can't find a solution to your problem, please contact our support team. Include detailed information about the issue you're experiencing for faster resolution.

Release Notes

Stay up to date with the latest updates and improvements to our platform.

Version 2.8.0 - June 2024
New Features
  • Added support for macrocycle ADME prediction in ForceADME
  • Introduced advanced binding site prediction in GeinDock
  • Released beta version of ForcePhrase for literature mining
  • Added batch processing for AI Detection tool
Improvements
  • Enhanced LogP prediction accuracy with new consensus algorithm
  • Improved GeinDock scoring function for fragment-like compounds
  • Optimized API performance with 2x throughput increase
  • Updated user interface for better mobile experience
Bug Fixes
  • Fixed rare issue with aromatic ring detection in certain heterocycles
  • Resolved API timeout on large batch submissions
  • Fixed sorting in analysis history view
  • Corrected minor display issues in Safari browser
Version 2.7.2 - April 2024
Improvements
  • Updated molecular visualization engine for better performance
  • Added 200+ new protein structures to the docking database
  • Enhanced report generation with new visualization options
Bug Fixes
  • Fixed issue with export to Excel on certain property sets
  • Resolved authentication issue for some API clients
  • Fixed inconsistent behavior in JSME molecular editor
View All Release Notes